From a3ff1940e93d57b74e143b36423b320d8677ef5b Mon Sep 17 00:00:00 2001 From: Colin Kealty <3266127+bartowski1182@users.noreply.github.com> Date: Wed, 4 Mar 2026 13:42:59 -0500 Subject: [PATCH] Fix merge conflicts, add more schemas --- tests/CMakeLists.txt | 6 + tests/gguf-model-data.cpp | 1 + tests/snapshots/deepseek-v3.1.schema | 1956 +++++++++ tests/snapshots/gemma-3-4b-it.schema | 1455 ++++++ tests/snapshots/gpt-oss-120b.schema | 1616 +++++++ .../meta-llama-3.1-70b-instruct.schema | 3899 +++++++++++++++++ .../snapshots/nemotron-nano-3-30b-a3b.schema | 696 +++ tests/snapshots/qwen3-14b.schema | 1908 ++++++++ tests/snapshots/qwen3-coder-next.schema | 1713 ++++++++ tests/snapshots/qwen3.5-27b.schema | 1837 ++++++++ tests/snapshots/qwen3.5-397b-a17b.schema | 2148 +++++++++ tests/snapshots/step-3.5-flash.schema | 2453 +++++++++++ tests/test-quant-type-selection.cpp | 2 +- 13 files changed, 19689 insertions(+), 1 deletion(-) create mode 100644 tests/snapshots/deepseek-v3.1.schema create mode 100644 tests/snapshots/gemma-3-4b-it.schema create mode 100644 tests/snapshots/gpt-oss-120b.schema create mode 100644 tests/snapshots/meta-llama-3.1-70b-instruct.schema create mode 100644 tests/snapshots/nemotron-nano-3-30b-a3b.schema create mode 100644 tests/snapshots/qwen3-14b.schema create mode 100644 tests/snapshots/qwen3-coder-next.schema create mode 100644 tests/snapshots/qwen3.5-27b.schema create mode 100644 tests/snapshots/qwen3.5-397b-a17b.schema create mode 100644 tests/snapshots/step-3.5-flash.schema diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt index 9582164b58..ced1f3943c 100644 --- a/tests/CMakeLists.txt +++ b/tests/CMakeLists.txt @@ -274,6 +274,12 @@ if (TARGET cpp-httplib) add_executable(test-gguf-model-data test-gguf-model-data.cpp) target_link_libraries(test-gguf-model-data PRIVATE gguf-model-data common) llama_test(test-gguf-model-data LABEL "model") + + # test-quant-type-selection requires gguf-model-data for remote model metadata + llama_build_and_test(test-quant-type-selection.cpp LABEL "model") + target_link_libraries(test-quant-type-selection PRIVATE gguf-model-data) + target_compile_definitions(test-quant-type-selection PRIVATE + SNAPSHOT_DIR="${CMAKE_CURRENT_SOURCE_DIR}/snapshots") endif() endif() diff --git a/tests/gguf-model-data.cpp b/tests/gguf-model-data.cpp index aa550c9538..97de083568 100644 --- a/tests/gguf-model-data.cpp +++ b/tests/gguf-model-data.cpp @@ -151,6 +151,7 @@ static bool gguf_read_uint32_val(gguf_buf_reader & r, int32_t vtype, uint32_t & return false; } } + return true; } if (vtype == GGUF_TYPE_UINT8) { uint8_t v; diff --git a/tests/snapshots/deepseek-v3.1.schema b/tests/snapshots/deepseek-v3.1.schema new file mode 100644 index 0000000000..a5bce29b27 --- /dev/null +++ b/tests/snapshots/deepseek-v3.1.schema @@ -0,0 +1,1956 @@ +# Model: DeepSeek-V3.1 +# n_embd=7168, n_ff=18432, n_vocab=129280, n_layer=61, n_head=128, n_head_kv=1, n_expert=256 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q6_K +blk.0.attn_output.weight q3_K +blk.0.ffn_down.weight q3_K +blk.1.attn_output.weight q3_K +blk.1.ffn_down.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K +blk.4.attn_output.weight q3_K +blk.4.ffn_down_exps.weight q3_K +blk.4.ffn_down_shexp.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.attn_output.weight q3_K +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.attn_output.weight q3_K +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.attn_output.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.attn_output.weight q3_K +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.attn_output.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.attn_output.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.attn_output.weight q3_K +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.attn_output.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.attn_output.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.attn_output.weight q3_K +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.attn_output.weight q3_K +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_output.weight q3_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.attn_output.weight q3_K +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.45.attn_output.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.46.attn_output.weight q3_K +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.attn_output.weight q3_K +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.48.attn_output.weight q3_K +blk.48.ffn_down_exps.weight q3_K +blk.48.ffn_down_shexp.weight q3_K +blk.49.attn_output.weight q3_K +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.50.attn_output.weight q3_K +blk.50.ffn_down_exps.weight q3_K +blk.50.ffn_down_shexp.weight q3_K +blk.51.attn_output.weight q3_K +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K +blk.52.attn_output.weight q3_K +blk.52.ffn_down_exps.weight q3_K +blk.52.ffn_down_shexp.weight q3_K +blk.53.attn_output.weight q3_K +blk.53.ffn_down_exps.weight q3_K +blk.53.ffn_down_shexp.weight q3_K +blk.54.attn_output.weight q3_K +blk.54.ffn_down_exps.weight q3_K +blk.54.ffn_down_shexp.weight q3_K +blk.55.attn_output.weight q3_K +blk.55.ffn_down_exps.weight q3_K +blk.55.ffn_down_shexp.weight q3_K +blk.56.attn_output.weight q3_K +blk.56.ffn_down_exps.weight q3_K +blk.56.ffn_down_shexp.weight q3_K +blk.57.attn_output.weight q3_K +blk.57.ffn_down_exps.weight q3_K +blk.57.ffn_down_shexp.weight q3_K +blk.58.attn_output.weight q3_K +blk.58.ffn_down_exps.weight q3_K +blk.58.ffn_down_shexp.weight q3_K +blk.59.attn_output.weight q3_K +blk.59.ffn_down_exps.weight q3_K +blk.59.ffn_down_shexp.weight q3_K +blk.60.attn_output.weight q3_K +blk.60.ffn_down_exps.weight q3_K +blk.60.ffn_down_shexp.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q4_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.ffn_down_exps.weight q4_K +blk.7.ffn_down_shexp.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.8.ffn_down_shexp.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.ffn_down_exps.weight q4_K +blk.9.ffn_down_shexp.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.10.ffn_down_shexp.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.ffn_down_exps.weight q4_K +blk.11.ffn_down_shexp.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.ffn_down_exps.weight q4_K +blk.12.ffn_down_shexp.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.ffn_down_exps.weight q4_K +blk.13.ffn_down_shexp.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.ffn_down_exps.weight q4_K +blk.14.ffn_down_shexp.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.ffn_down_exps.weight q4_K +blk.16.ffn_down_shexp.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.17.ffn_down_shexp.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.ffn_down_exps.weight q4_K +blk.18.ffn_down_shexp.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.ffn_down_exps.weight q4_K +blk.19.ffn_down_shexp.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.20.ffn_down_shexp.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.ffn_down_exps.weight q4_K +blk.21.ffn_down_shexp.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.22.ffn_down_shexp.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.23.ffn_down_shexp.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.24.ffn_down_shexp.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.ffn_down_exps.weight q4_K +blk.25.ffn_down_shexp.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.ffn_down_exps.weight q4_K +blk.26.ffn_down_shexp.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.ffn_down_exps.weight q4_K +blk.28.ffn_down_shexp.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.29.ffn_down_shexp.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.ffn_down_exps.weight q4_K +blk.30.ffn_down_shexp.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.ffn_down_exps.weight q4_K +blk.32.ffn_down_shexp.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.ffn_down_exps.weight q4_K +blk.33.ffn_down_shexp.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.34.ffn_down_shexp.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.ffn_down_exps.weight q4_K +blk.35.ffn_down_shexp.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.ffn_down_exps.weight q4_K +blk.36.ffn_down_shexp.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.ffn_down_exps.weight q4_K +blk.37.ffn_down_shexp.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.ffn_down_exps.weight q4_K +blk.38.ffn_down_shexp.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.ffn_down_exps.weight q4_K +blk.39.ffn_down_shexp.weight q4_K +blk.40.attn_output.weight q4_K +blk.40.ffn_down_exps.weight q4_K +blk.40.ffn_down_shexp.weight q4_K +blk.41.attn_output.weight q4_K +blk.41.ffn_down_exps.weight q4_K +blk.41.ffn_down_shexp.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.ffn_down_exps.weight q4_K +blk.42.ffn_down_shexp.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K +blk.44.attn_output.weight q4_K +blk.44.ffn_down_exps.weight q4_K +blk.44.ffn_down_shexp.weight q4_K +blk.45.attn_output.weight q4_K +blk.45.ffn_down_exps.weight q4_K +blk.45.ffn_down_shexp.weight q4_K +blk.46.attn_output.weight q4_K +blk.46.ffn_down_exps.weight q4_K +blk.46.ffn_down_shexp.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.ffn_down_exps.weight q4_K +blk.47.ffn_down_shexp.weight q4_K +blk.48.attn_output.weight q4_K +blk.48.ffn_down_exps.weight q4_K +blk.48.ffn_down_shexp.weight q4_K +blk.49.attn_output.weight q4_K +blk.49.ffn_down_exps.weight q4_K +blk.49.ffn_down_shexp.weight q4_K +blk.50.attn_output.weight q4_K +blk.50.ffn_down_exps.weight q4_K +blk.50.ffn_down_shexp.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.ffn_down_exps.weight q4_K +blk.51.ffn_down_shexp.weight q4_K +blk.52.attn_output.weight q4_K +blk.52.ffn_down_exps.weight q4_K +blk.52.ffn_down_shexp.weight q4_K +blk.53.attn_output.weight q4_K +blk.53.ffn_down_exps.weight q4_K +blk.53.ffn_down_shexp.weight q4_K +blk.54.attn_output.weight q4_K +blk.54.ffn_down_exps.weight q4_K +blk.54.ffn_down_shexp.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.ffn_down_exps.weight q4_K +blk.55.ffn_down_shexp.weight q4_K +blk.56.attn_output.weight q4_K +blk.56.ffn_down_exps.weight q4_K +blk.56.ffn_down_shexp.weight q4_K +blk.57.attn_output.weight q4_K +blk.57.ffn_down_exps.weight q4_K +blk.57.ffn_down_shexp.weight q4_K +blk.58.attn_output.weight q4_K +blk.58.ffn_down_exps.weight q4_K +blk.58.ffn_down_shexp.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.ffn_down_exps.weight q4_K +blk.59.ffn_down_shexp.weight q4_K +blk.60.attn_output.weight q4_K +blk.60.ffn_down_exps.weight q4_K +blk.60.ffn_down_shexp.weight q4_K + +[Q3_K_L] q3_K +output.weight q6_K +blk.0.attn_output.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.ffn_down_exps.weight q5_K +blk.7.ffn_down_shexp.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.ffn_down_exps.weight q5_K +blk.8.ffn_down_shexp.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.ffn_down_exps.weight q5_K +blk.9.ffn_down_shexp.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.ffn_down_exps.weight q5_K +blk.10.ffn_down_shexp.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.ffn_down_exps.weight q5_K +blk.11.ffn_down_shexp.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.ffn_down_exps.weight q5_K +blk.12.ffn_down_shexp.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.ffn_down_exps.weight q5_K +blk.13.ffn_down_shexp.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.ffn_down_exps.weight q5_K +blk.14.ffn_down_shexp.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.ffn_down_exps.weight q5_K +blk.16.ffn_down_shexp.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.ffn_down_exps.weight q5_K +blk.17.ffn_down_shexp.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.ffn_down_exps.weight q5_K +blk.18.ffn_down_shexp.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.ffn_down_exps.weight q5_K +blk.19.ffn_down_shexp.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.ffn_down_exps.weight q5_K +blk.20.ffn_down_shexp.weight q5_K +blk.21.attn_output.weight q5_K +blk.21.ffn_down_exps.weight q5_K +blk.21.ffn_down_shexp.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.ffn_down_exps.weight q5_K +blk.22.ffn_down_shexp.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.23.ffn_down_shexp.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.ffn_down_exps.weight q5_K +blk.24.ffn_down_shexp.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.ffn_down_exps.weight q5_K +blk.25.ffn_down_shexp.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.ffn_down_exps.weight q5_K +blk.26.ffn_down_shexp.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.ffn_down_exps.weight q5_K +blk.28.ffn_down_shexp.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.ffn_down_exps.weight q5_K +blk.29.ffn_down_shexp.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.ffn_down_exps.weight q5_K +blk.30.ffn_down_shexp.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.ffn_down_exps.weight q5_K +blk.32.ffn_down_shexp.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.ffn_down_exps.weight q5_K +blk.33.ffn_down_shexp.weight q5_K +blk.34.attn_output.weight q5_K +blk.34.ffn_down_exps.weight q5_K +blk.34.ffn_down_shexp.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.ffn_down_exps.weight q5_K +blk.35.ffn_down_shexp.weight q5_K +blk.36.attn_output.weight q5_K +blk.36.ffn_down_exps.weight q5_K +blk.36.ffn_down_shexp.weight q5_K +blk.37.attn_output.weight q5_K +blk.37.ffn_down_exps.weight q5_K +blk.37.ffn_down_shexp.weight q5_K +blk.38.attn_output.weight q5_K +blk.38.ffn_down_exps.weight q5_K +blk.38.ffn_down_shexp.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.ffn_down_exps.weight q5_K +blk.39.ffn_down_shexp.weight q5_K +blk.40.attn_output.weight q5_K +blk.40.ffn_down_exps.weight q5_K +blk.40.ffn_down_shexp.weight q5_K +blk.41.attn_output.weight q5_K +blk.41.ffn_down_exps.weight q5_K +blk.41.ffn_down_shexp.weight q5_K +blk.42.attn_output.weight q5_K +blk.42.ffn_down_exps.weight q5_K +blk.42.ffn_down_shexp.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K +blk.44.attn_output.weight q5_K +blk.44.ffn_down_exps.weight q5_K +blk.44.ffn_down_shexp.weight q5_K +blk.45.attn_output.weight q5_K +blk.45.ffn_down_exps.weight q5_K +blk.45.ffn_down_shexp.weight q5_K +blk.46.attn_output.weight q5_K +blk.46.ffn_down_exps.weight q5_K +blk.46.ffn_down_shexp.weight q5_K +blk.47.attn_output.weight q5_K +blk.47.ffn_down_exps.weight q5_K +blk.47.ffn_down_shexp.weight q5_K +blk.48.attn_output.weight q5_K +blk.48.ffn_down_exps.weight q5_K +blk.48.ffn_down_shexp.weight q5_K +blk.49.attn_output.weight q5_K +blk.49.ffn_down_exps.weight q5_K +blk.49.ffn_down_shexp.weight q5_K +blk.50.attn_output.weight q5_K +blk.50.ffn_down_exps.weight q5_K +blk.50.ffn_down_shexp.weight q5_K +blk.51.attn_output.weight q5_K +blk.51.ffn_down_exps.weight q5_K +blk.51.ffn_down_shexp.weight q5_K +blk.52.attn_output.weight q5_K +blk.52.ffn_down_exps.weight q5_K +blk.52.ffn_down_shexp.weight q5_K +blk.53.attn_output.weight q5_K +blk.53.ffn_down_exps.weight q5_K +blk.53.ffn_down_shexp.weight q5_K +blk.54.attn_output.weight q5_K +blk.54.ffn_down_exps.weight q5_K +blk.54.ffn_down_shexp.weight q5_K +blk.55.attn_output.weight q5_K +blk.55.ffn_down_exps.weight q5_K +blk.55.ffn_down_shexp.weight q5_K +blk.56.attn_output.weight q5_K +blk.56.ffn_down_exps.weight q5_K +blk.56.ffn_down_shexp.weight q5_K +blk.57.attn_output.weight q5_K +blk.57.ffn_down_exps.weight q5_K +blk.57.ffn_down_shexp.weight q5_K +blk.58.attn_output.weight q5_K +blk.58.ffn_down_exps.weight q5_K +blk.58.ffn_down_shexp.weight q5_K +blk.59.attn_output.weight q5_K +blk.59.ffn_down_exps.weight q5_K +blk.59.ffn_down_shexp.weight q5_K +blk.60.attn_output.weight q5_K +blk.60.ffn_down_exps.weight q5_K +blk.60.ffn_down_shexp.weight q5_K + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K + +[Q4_K_M] q4_K +output.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K +blk.6.ffn_down_exps.weight q6_K +blk.6.ffn_down_shexp.weight q6_K +blk.9.ffn_down_exps.weight q6_K +blk.9.ffn_down_shexp.weight q6_K +blk.12.ffn_down_exps.weight q6_K +blk.12.ffn_down_shexp.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.15.ffn_down_shexp.weight q6_K +blk.18.ffn_down_exps.weight q6_K +blk.18.ffn_down_shexp.weight q6_K +blk.21.ffn_down_exps.weight q6_K +blk.21.ffn_down_shexp.weight q6_K +blk.24.ffn_down_exps.weight q6_K +blk.24.ffn_down_shexp.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.27.ffn_down_shexp.weight q6_K +blk.30.ffn_down_exps.weight q6_K +blk.30.ffn_down_shexp.weight q6_K +blk.33.ffn_down_exps.weight q6_K +blk.33.ffn_down_shexp.weight q6_K +blk.36.ffn_down_exps.weight q6_K +blk.36.ffn_down_shexp.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.48.ffn_down_exps.weight q6_K +blk.48.ffn_down_shexp.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K +blk.53.ffn_down_exps.weight q6_K +blk.53.ffn_down_shexp.weight q6_K +blk.54.ffn_down_exps.weight q6_K +blk.54.ffn_down_shexp.weight q6_K +blk.55.ffn_down_exps.weight q6_K +blk.55.ffn_down_shexp.weight q6_K +blk.56.ffn_down_exps.weight q6_K +blk.56.ffn_down_shexp.weight q6_K +blk.57.ffn_down_exps.weight q6_K +blk.57.ffn_down_shexp.weight q6_K +blk.58.ffn_down_exps.weight q6_K +blk.58.ffn_down_shexp.weight q6_K +blk.59.ffn_down_exps.weight q6_K +blk.59.ffn_down_shexp.weight q6_K +blk.60.ffn_down_exps.weight q6_K +blk.60.ffn_down_shexp.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +output.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K +blk.6.ffn_down_exps.weight q6_K +blk.6.ffn_down_shexp.weight q6_K +blk.9.ffn_down_exps.weight q6_K +blk.9.ffn_down_shexp.weight q6_K +blk.12.ffn_down_exps.weight q6_K +blk.12.ffn_down_shexp.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.15.ffn_down_shexp.weight q6_K +blk.18.ffn_down_exps.weight q6_K +blk.18.ffn_down_shexp.weight q6_K +blk.21.ffn_down_exps.weight q6_K +blk.21.ffn_down_shexp.weight q6_K +blk.24.ffn_down_exps.weight q6_K +blk.24.ffn_down_shexp.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.27.ffn_down_shexp.weight q6_K +blk.30.ffn_down_exps.weight q6_K +blk.30.ffn_down_shexp.weight q6_K +blk.33.ffn_down_exps.weight q6_K +blk.33.ffn_down_shexp.weight q6_K +blk.36.ffn_down_exps.weight q6_K +blk.36.ffn_down_shexp.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.48.ffn_down_exps.weight q6_K +blk.48.ffn_down_shexp.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K +blk.53.ffn_down_exps.weight q6_K +blk.53.ffn_down_shexp.weight q6_K +blk.54.ffn_down_exps.weight q6_K +blk.54.ffn_down_shexp.weight q6_K +blk.55.ffn_down_exps.weight q6_K +blk.55.ffn_down_shexp.weight q6_K +blk.56.ffn_down_exps.weight q6_K +blk.56.ffn_down_shexp.weight q6_K +blk.57.ffn_down_exps.weight q6_K +blk.57.ffn_down_shexp.weight q6_K +blk.58.ffn_down_exps.weight q6_K +blk.58.ffn_down_shexp.weight q6_K +blk.59.ffn_down_exps.weight q6_K +blk.59.ffn_down_shexp.weight q6_K +blk.60.ffn_down_exps.weight q6_K +blk.60.ffn_down_shexp.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.ffn_down_exps.weight q2_K +blk.4.ffn_down_shexp.weight q2_K + +[IQ2_XS] iq2_xs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.ffn_down_exps.weight q2_K +blk.4.ffn_down_shexp.weight q2_K + +[Q2_K_S] q2_K +output.weight q6_K +blk.0.ffn_down.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K + +[IQ3_XS] iq3_s +output.weight q6_K +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.9.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.12.ffn_up_shexp.weight iq3_xxs +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_gate_shexp.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs +blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs +blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs +blk.32.ffn_up_exps.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs +blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs +blk.33.ffn_up_exps.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs +blk.34.ffn_gate_exps.weight iq3_xxs +blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs +blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs +blk.35.ffn_up_exps.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs +blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs +blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs +blk.37.ffn_up_exps.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs +blk.39.ffn_gate_exps.weight iq3_xxs +blk.39.ffn_gate_shexp.weight iq3_xxs +blk.39.ffn_up_exps.weight iq3_xxs +blk.39.ffn_up_shexp.weight iq3_xxs +blk.40.ffn_gate_exps.weight iq3_xxs +blk.40.ffn_gate_shexp.weight iq3_xxs +blk.40.ffn_up_exps.weight iq3_xxs +blk.40.ffn_up_shexp.weight iq3_xxs +blk.41.ffn_gate_exps.weight iq3_xxs +blk.41.ffn_gate_shexp.weight iq3_xxs +blk.41.ffn_up_exps.weight iq3_xxs +blk.41.ffn_up_shexp.weight iq3_xxs +blk.42.ffn_gate_exps.weight iq3_xxs +blk.42.ffn_gate_shexp.weight iq3_xxs +blk.42.ffn_up_exps.weight iq3_xxs +blk.42.ffn_up_shexp.weight iq3_xxs +blk.43.ffn_gate_exps.weight iq3_xxs +blk.43.ffn_gate_shexp.weight iq3_xxs +blk.43.ffn_up_exps.weight iq3_xxs +blk.43.ffn_up_shexp.weight iq3_xxs +blk.44.ffn_gate_exps.weight iq3_xxs +blk.44.ffn_gate_shexp.weight iq3_xxs +blk.44.ffn_up_exps.weight iq3_xxs +blk.44.ffn_up_shexp.weight iq3_xxs +blk.45.ffn_gate_exps.weight iq3_xxs +blk.45.ffn_gate_shexp.weight iq3_xxs +blk.45.ffn_up_exps.weight iq3_xxs +blk.45.ffn_up_shexp.weight iq3_xxs +blk.46.ffn_gate_exps.weight iq3_xxs +blk.46.ffn_gate_shexp.weight iq3_xxs +blk.46.ffn_up_exps.weight iq3_xxs +blk.46.ffn_up_shexp.weight iq3_xxs +blk.47.ffn_gate_exps.weight iq3_xxs +blk.47.ffn_gate_shexp.weight iq3_xxs +blk.47.ffn_up_exps.weight iq3_xxs +blk.47.ffn_up_shexp.weight iq3_xxs +blk.48.ffn_gate_exps.weight iq3_xxs +blk.48.ffn_gate_shexp.weight iq3_xxs +blk.48.ffn_up_exps.weight iq3_xxs +blk.48.ffn_up_shexp.weight iq3_xxs +blk.49.ffn_gate_exps.weight iq3_xxs +blk.49.ffn_gate_shexp.weight iq3_xxs +blk.49.ffn_up_exps.weight iq3_xxs +blk.49.ffn_up_shexp.weight iq3_xxs +blk.50.ffn_gate_exps.weight iq3_xxs +blk.50.ffn_gate_shexp.weight iq3_xxs +blk.50.ffn_up_exps.weight iq3_xxs +blk.50.ffn_up_shexp.weight iq3_xxs +blk.51.ffn_gate_exps.weight iq3_xxs +blk.51.ffn_gate_shexp.weight iq3_xxs +blk.51.ffn_up_exps.weight iq3_xxs +blk.51.ffn_up_shexp.weight iq3_xxs +blk.52.ffn_gate_exps.weight iq3_xxs +blk.52.ffn_gate_shexp.weight iq3_xxs +blk.52.ffn_up_exps.weight iq3_xxs +blk.52.ffn_up_shexp.weight iq3_xxs + +[IQ3_XXS] iq3_xxs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.ffn_down.weight q4_K +blk.1.attn_output.weight iq3_s +blk.1.ffn_down.weight q4_K +blk.2.attn_output.weight iq3_s +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight iq3_s +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_output.weight iq3_s +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.attn_output.weight iq3_s +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.attn_output.weight iq3_s +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.attn_output.weight iq3_s +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_output.weight iq3_s +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.attn_output.weight iq3_s +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.attn_output.weight iq3_s +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.attn_output.weight iq3_s +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_output.weight iq3_s +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.attn_output.weight iq3_s +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.attn_output.weight iq3_s +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.attn_output.weight iq3_s +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight iq3_s +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.attn_output.weight iq3_s +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.attn_output.weight iq3_s +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.attn_output.weight iq3_s +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_output.weight iq3_s +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.attn_output.weight iq3_s +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.attn_output.weight iq3_s +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.attn_output.weight iq3_s +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_output.weight iq3_s +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.attn_output.weight iq3_s +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.attn_output.weight iq3_s +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.attn_output.weight iq3_s +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_output.weight iq3_s +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.attn_output.weight iq3_s +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.attn_output.weight iq3_s +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.attn_output.weight iq3_s +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_output.weight iq3_s +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.attn_output.weight iq3_s +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.attn_output.weight iq3_s +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.attn_output.weight iq3_s +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_output.weight iq3_s +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.attn_output.weight iq3_s +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.attn_output.weight iq3_s +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.attn_output.weight iq3_s +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_output.weight iq3_s +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.attn_output.weight iq3_s +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.45.attn_output.weight iq3_s +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.46.attn_output.weight iq3_s +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.attn_output.weight iq3_s +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.48.attn_output.weight iq3_s +blk.48.ffn_down_exps.weight q3_K +blk.48.ffn_down_shexp.weight q3_K +blk.49.attn_output.weight iq3_s +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.50.attn_output.weight iq3_s +blk.50.ffn_down_exps.weight q3_K +blk.50.ffn_down_shexp.weight q3_K +blk.51.attn_output.weight iq3_s +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K +blk.52.attn_output.weight iq3_s +blk.52.ffn_down_exps.weight q3_K +blk.52.ffn_down_shexp.weight q3_K +blk.53.attn_output.weight iq3_s +blk.53.ffn_down_exps.weight q3_K +blk.53.ffn_down_shexp.weight q3_K +blk.54.attn_output.weight iq3_s +blk.54.ffn_down_exps.weight q3_K +blk.54.ffn_down_shexp.weight q3_K +blk.55.attn_output.weight iq3_s +blk.55.ffn_down_exps.weight q3_K +blk.55.ffn_down_shexp.weight q3_K +blk.56.attn_output.weight iq3_s +blk.56.ffn_down_exps.weight q3_K +blk.56.ffn_down_shexp.weight q3_K +blk.57.attn_output.weight iq3_s +blk.57.ffn_down_exps.weight q3_K +blk.57.ffn_down_shexp.weight q3_K +blk.58.attn_output.weight iq3_s +blk.58.ffn_down_exps.weight q3_K +blk.58.ffn_down_shexp.weight q3_K +blk.59.attn_output.weight iq3_s +blk.59.ffn_down_exps.weight q3_K +blk.59.ffn_down_shexp.weight q3_K +blk.60.attn_output.weight iq3_s +blk.60.ffn_down_exps.weight q3_K +blk.60.ffn_down_shexp.weight q3_K + +[IQ1_S] iq1_s +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.ffn_down_exps.weight q2_K +blk.4.ffn_down_shexp.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.6.attn_output.weight iq2_xxs +blk.7.attn_output.weight iq2_xxs +blk.8.attn_output.weight iq2_xxs +blk.9.attn_output.weight iq2_xxs +blk.10.attn_output.weight iq2_xxs +blk.11.attn_output.weight iq2_xxs +blk.12.attn_output.weight iq2_xxs +blk.13.attn_output.weight iq2_xxs +blk.14.attn_output.weight iq2_xxs +blk.15.attn_output.weight iq2_xxs +blk.16.attn_output.weight iq2_xxs +blk.17.attn_output.weight iq2_xxs +blk.18.attn_output.weight iq2_xxs +blk.19.attn_output.weight iq2_xxs +blk.20.attn_output.weight iq2_xxs +blk.21.attn_output.weight iq2_xxs +blk.22.attn_output.weight iq2_xxs +blk.23.attn_output.weight iq2_xxs +blk.24.attn_output.weight iq2_xxs +blk.25.attn_output.weight iq2_xxs +blk.26.attn_output.weight iq2_xxs +blk.27.attn_output.weight iq2_xxs +blk.28.attn_output.weight iq2_xxs +blk.29.attn_output.weight iq2_xxs +blk.30.attn_output.weight iq2_xxs +blk.31.attn_output.weight iq2_xxs +blk.32.attn_output.weight iq2_xxs +blk.33.attn_output.weight iq2_xxs +blk.34.attn_output.weight iq2_xxs +blk.35.attn_output.weight iq2_xxs +blk.36.attn_output.weight iq2_xxs +blk.37.attn_output.weight iq2_xxs +blk.38.attn_output.weight iq2_xxs +blk.39.attn_output.weight iq2_xxs +blk.40.attn_output.weight iq2_xxs +blk.41.attn_output.weight iq2_xxs +blk.42.attn_output.weight iq2_xxs +blk.43.attn_output.weight iq2_xxs +blk.44.attn_output.weight iq2_xxs +blk.45.attn_output.weight iq2_xxs +blk.46.attn_output.weight iq2_xxs +blk.47.attn_output.weight iq2_xxs +blk.48.attn_output.weight iq2_xxs +blk.49.attn_output.weight iq2_xxs +blk.50.attn_output.weight iq2_xxs +blk.51.attn_output.weight iq2_xxs +blk.52.attn_output.weight iq2_xxs +blk.53.attn_output.weight iq2_xxs +blk.54.attn_output.weight iq2_xxs +blk.55.attn_output.weight iq2_xxs +blk.56.attn_output.weight iq2_xxs +blk.57.attn_output.weight iq2_xxs +blk.58.attn_output.weight iq2_xxs +blk.59.attn_output.weight iq2_xxs +blk.60.attn_output.weight iq2_xxs + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K + +[IQ3_S] iq3_s +output.weight q6_K + +[IQ3_M] iq3_s +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_output.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.8.attn_output.weight q4_K +blk.9.attn_output.weight q4_K +blk.10.attn_output.weight q4_K +blk.11.attn_output.weight q4_K +blk.12.attn_output.weight q4_K +blk.13.attn_output.weight q4_K +blk.14.attn_output.weight q4_K +blk.15.attn_output.weight q4_K +blk.16.attn_output.weight q4_K +blk.17.attn_output.weight q4_K +blk.18.attn_output.weight q4_K +blk.19.attn_output.weight q4_K +blk.20.attn_output.weight q4_K +blk.21.attn_output.weight q4_K +blk.22.attn_output.weight q4_K +blk.23.attn_output.weight q4_K +blk.24.attn_output.weight q4_K +blk.25.attn_output.weight q4_K +blk.26.attn_output.weight q4_K +blk.27.attn_output.weight q4_K +blk.28.attn_output.weight q4_K +blk.29.attn_output.weight q4_K +blk.30.attn_output.weight q4_K +blk.31.attn_output.weight q4_K +blk.32.attn_output.weight q4_K +blk.33.attn_output.weight q4_K +blk.34.attn_output.weight q4_K +blk.35.attn_output.weight q4_K +blk.36.attn_output.weight q4_K +blk.37.attn_output.weight q4_K +blk.38.attn_output.weight q4_K +blk.39.attn_output.weight q4_K +blk.40.attn_output.weight q4_K +blk.41.attn_output.weight q4_K +blk.42.attn_output.weight q4_K +blk.43.attn_output.weight q4_K +blk.44.attn_output.weight q4_K +blk.45.attn_output.weight q4_K +blk.46.attn_output.weight q4_K +blk.47.attn_output.weight q4_K +blk.48.attn_output.weight q4_K +blk.49.attn_output.weight q4_K +blk.50.attn_output.weight q4_K +blk.51.attn_output.weight q4_K +blk.52.attn_output.weight q4_K +blk.53.attn_output.weight q4_K +blk.54.attn_output.weight q4_K +blk.55.attn_output.weight q4_K +blk.56.attn_output.weight q4_K +blk.57.attn_output.weight q4_K +blk.58.attn_output.weight q4_K +blk.59.attn_output.weight q4_K +blk.60.attn_output.weight q4_K + +[IQ2_S] iq2_xs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.ffn_down_exps.weight iq3_s +blk.4.ffn_down_shexp.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.15.attn_output.weight iq3_s +blk.16.attn_output.weight iq3_s +blk.17.attn_output.weight iq3_s +blk.18.attn_output.weight iq3_s +blk.19.attn_output.weight iq3_s +blk.20.attn_output.weight iq3_s +blk.21.attn_output.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.23.attn_output.weight iq3_s +blk.24.attn_output.weight iq3_s +blk.25.attn_output.weight iq3_s +blk.26.attn_output.weight iq3_s +blk.27.attn_output.weight iq3_s +blk.28.attn_output.weight iq3_s +blk.29.attn_output.weight iq3_s +blk.30.attn_output.weight iq3_s +blk.31.attn_output.weight iq3_s +blk.32.attn_output.weight iq3_s +blk.33.attn_output.weight iq3_s +blk.34.attn_output.weight iq3_s +blk.35.attn_output.weight iq3_s +blk.36.attn_output.weight iq3_s +blk.37.attn_output.weight iq3_s +blk.38.attn_output.weight iq3_s +blk.39.attn_output.weight iq3_s +blk.40.attn_output.weight iq3_s +blk.41.attn_output.weight iq3_s +blk.42.attn_output.weight iq3_s +blk.43.attn_output.weight iq3_s +blk.44.attn_output.weight iq3_s +blk.45.attn_output.weight iq3_s +blk.46.attn_output.weight iq3_s +blk.47.attn_output.weight iq3_s +blk.48.attn_output.weight iq3_s +blk.49.attn_output.weight iq3_s +blk.50.attn_output.weight iq3_s +blk.51.attn_output.weight iq3_s +blk.52.attn_output.weight iq3_s +blk.53.attn_output.weight iq3_s +blk.54.attn_output.weight iq3_s +blk.55.attn_output.weight iq3_s +blk.56.attn_output.weight iq3_s +blk.57.attn_output.weight iq3_s +blk.58.attn_output.weight iq3_s +blk.59.attn_output.weight iq3_s +blk.60.attn_output.weight iq3_s + +[IQ2_M] iq2_s +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.ffn_down_exps.weight iq3_s +blk.4.ffn_down_shexp.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.15.attn_output.weight iq3_s +blk.16.attn_output.weight iq3_s +blk.17.attn_output.weight iq3_s +blk.18.attn_output.weight iq3_s +blk.19.attn_output.weight iq3_s +blk.20.attn_output.weight iq3_s +blk.21.attn_output.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.23.attn_output.weight iq3_s +blk.24.attn_output.weight iq3_s +blk.25.attn_output.weight iq3_s +blk.26.attn_output.weight iq3_s +blk.27.attn_output.weight iq3_s +blk.28.attn_output.weight iq3_s +blk.29.attn_output.weight iq3_s +blk.30.attn_output.weight iq3_s +blk.31.attn_output.weight iq3_s +blk.32.attn_output.weight iq3_s +blk.33.attn_output.weight iq3_s +blk.34.attn_output.weight iq3_s +blk.35.attn_output.weight iq3_s +blk.36.attn_output.weight iq3_s +blk.37.attn_output.weight iq3_s +blk.38.attn_output.weight iq3_s +blk.39.attn_output.weight iq3_s +blk.40.attn_output.weight iq3_s +blk.41.attn_output.weight iq3_s +blk.42.attn_output.weight iq3_s +blk.43.attn_output.weight iq3_s +blk.44.attn_output.weight iq3_s +blk.45.attn_output.weight iq3_s +blk.46.attn_output.weight iq3_s +blk.47.attn_output.weight iq3_s +blk.48.attn_output.weight iq3_s +blk.49.attn_output.weight iq3_s +blk.50.attn_output.weight iq3_s +blk.51.attn_output.weight iq3_s +blk.52.attn_output.weight iq3_s +blk.53.attn_output.weight iq3_s +blk.54.attn_output.weight iq3_s +blk.55.attn_output.weight iq3_s +blk.56.attn_output.weight iq3_s +blk.57.attn_output.weight iq3_s +blk.58.attn_output.weight iq3_s +blk.59.attn_output.weight iq3_s +blk.60.attn_output.weight iq3_s + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.ffn_down_exps.weight q2_K +blk.4.ffn_down_shexp.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.6.attn_output.weight iq2_xxs +blk.7.attn_output.weight iq2_xxs +blk.8.attn_output.weight iq2_xxs +blk.9.attn_output.weight iq2_xxs +blk.10.attn_output.weight iq2_xxs +blk.11.attn_output.weight iq2_xxs +blk.12.attn_output.weight iq2_xxs +blk.13.attn_output.weight iq2_xxs +blk.14.attn_output.weight iq2_xxs +blk.15.attn_output.weight iq2_xxs +blk.16.attn_output.weight iq2_xxs +blk.17.attn_output.weight iq2_xxs +blk.18.attn_output.weight iq2_xxs +blk.19.attn_output.weight iq2_xxs +blk.20.attn_output.weight iq2_xxs +blk.21.attn_output.weight iq2_xxs +blk.22.attn_output.weight iq2_xxs +blk.23.attn_output.weight iq2_xxs +blk.24.attn_output.weight iq2_xxs +blk.25.attn_output.weight iq2_xxs +blk.26.attn_output.weight iq2_xxs +blk.27.attn_output.weight iq2_xxs +blk.28.attn_output.weight iq2_xxs +blk.29.attn_output.weight iq2_xxs +blk.30.attn_output.weight iq2_xxs +blk.31.attn_output.weight iq2_xxs +blk.32.attn_output.weight iq2_xxs +blk.33.attn_output.weight iq2_xxs +blk.34.attn_output.weight iq2_xxs +blk.35.attn_output.weight iq2_xxs +blk.36.attn_output.weight iq2_xxs +blk.37.attn_output.weight iq2_xxs +blk.38.attn_output.weight iq2_xxs +blk.39.attn_output.weight iq2_xxs +blk.40.attn_output.weight iq2_xxs +blk.41.attn_output.weight iq2_xxs +blk.42.attn_output.weight iq2_xxs +blk.43.attn_output.weight iq2_xxs +blk.44.attn_output.weight iq2_xxs +blk.45.attn_output.weight iq2_xxs +blk.46.attn_output.weight iq2_xxs +blk.47.attn_output.weight iq2_xxs +blk.48.attn_output.weight iq2_xxs +blk.49.attn_output.weight iq2_xxs +blk.50.attn_output.weight iq2_xxs +blk.51.attn_output.weight iq2_xxs +blk.52.attn_output.weight iq2_xxs +blk.53.attn_output.weight iq2_xxs +blk.54.attn_output.weight iq2_xxs +blk.55.attn_output.weight iq2_xxs +blk.56.attn_output.weight iq2_xxs +blk.57.attn_output.weight iq2_xxs +blk.58.attn_output.weight iq2_xxs +blk.59.attn_output.weight iq2_xxs +blk.60.attn_output.weight iq2_xxs + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q6_K +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q6_K +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_kv_a_mqa.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q_a.weight q8_0 +blk.0.attn_q_b.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.1.attn_kv_a_mqa.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q_a.weight q8_0 +blk.1.attn_q_b.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.2.attn_kv_a_mqa.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q_a.weight q8_0 +blk.2.attn_q_b.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.3.attn_kv_a_mqa.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q_a.weight q8_0 +blk.3.attn_q_b.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 +blk.4.attn_kv_a_mqa.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q_a.weight q8_0 +blk.4.attn_q_b.weight q8_0 +blk.4.ffn_down_shexp.weight q8_0 +blk.4.ffn_gate_shexp.weight q8_0 +blk.4.ffn_up_shexp.weight q8_0 +blk.5.attn_kv_a_mqa.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q_a.weight q8_0 +blk.5.attn_q_b.weight q8_0 +blk.5.ffn_down_shexp.weight q8_0 +blk.5.ffn_gate_shexp.weight q8_0 +blk.5.ffn_up_shexp.weight q8_0 +blk.6.attn_kv_a_mqa.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q_a.weight q8_0 +blk.6.attn_q_b.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_gate_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 +blk.7.attn_kv_a_mqa.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q_a.weight q8_0 +blk.7.attn_q_b.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 +blk.8.attn_kv_a_mqa.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q_a.weight q8_0 +blk.8.attn_q_b.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_gate_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 +blk.9.attn_kv_a_mqa.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q_a.weight q8_0 +blk.9.attn_q_b.weight q8_0 +blk.9.ffn_down_shexp.weight q8_0 +blk.9.ffn_gate_shexp.weight q8_0 +blk.9.ffn_up_shexp.weight q8_0 +blk.10.attn_kv_a_mqa.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q_a.weight q8_0 +blk.10.attn_q_b.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_gate_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 +blk.11.attn_kv_a_mqa.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q_a.weight q8_0 +blk.11.attn_q_b.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 +blk.12.attn_kv_a_mqa.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q_a.weight q8_0 +blk.12.attn_q_b.weight q8_0 +blk.12.ffn_down_shexp.weight q8_0 +blk.12.ffn_gate_shexp.weight q8_0 +blk.12.ffn_up_shexp.weight q8_0 +blk.13.attn_kv_a_mqa.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q_a.weight q8_0 +blk.13.attn_q_b.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_gate_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 +blk.14.attn_kv_a_mqa.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q_a.weight q8_0 +blk.14.attn_q_b.weight q8_0 +blk.14.ffn_down_shexp.weight q8_0 +blk.14.ffn_gate_shexp.weight q8_0 +blk.14.ffn_up_shexp.weight q8_0 +blk.15.attn_kv_a_mqa.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q_a.weight q8_0 +blk.15.attn_q_b.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 +blk.16.attn_kv_a_mqa.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q_a.weight q8_0 +blk.16.attn_q_b.weight q8_0 +blk.16.ffn_down_shexp.weight q8_0 +blk.16.ffn_gate_shexp.weight q8_0 +blk.16.ffn_up_shexp.weight q8_0 +blk.17.attn_kv_a_mqa.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q_a.weight q8_0 +blk.17.attn_q_b.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_gate_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 +blk.18.attn_kv_a_mqa.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q_a.weight q8_0 +blk.18.attn_q_b.weight q8_0 +blk.18.ffn_down_shexp.weight q8_0 +blk.18.ffn_gate_shexp.weight q8_0 +blk.18.ffn_up_shexp.weight q8_0 +blk.19.attn_kv_a_mqa.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q_a.weight q8_0 +blk.19.attn_q_b.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 +blk.20.attn_kv_a_mqa.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q_a.weight q8_0 +blk.20.attn_q_b.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_gate_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 +blk.21.attn_kv_a_mqa.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.21.attn_q_a.weight q8_0 +blk.21.attn_q_b.weight q8_0 +blk.21.ffn_down_shexp.weight q8_0 +blk.21.ffn_gate_shexp.weight q8_0 +blk.21.ffn_up_shexp.weight q8_0 +blk.22.attn_kv_a_mqa.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q_a.weight q8_0 +blk.22.attn_q_b.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_gate_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 +blk.23.attn_kv_a_mqa.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q_a.weight q8_0 +blk.23.attn_q_b.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 +blk.24.attn_kv_a_mqa.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q_a.weight q8_0 +blk.24.attn_q_b.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_gate_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 +blk.25.attn_kv_a_mqa.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q_a.weight q8_0 +blk.25.attn_q_b.weight q8_0 +blk.25.ffn_down_shexp.weight q8_0 +blk.25.ffn_gate_shexp.weight q8_0 +blk.25.ffn_up_shexp.weight q8_0 +blk.26.attn_kv_a_mqa.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q_a.weight q8_0 +blk.26.attn_q_b.weight q8_0 +blk.26.ffn_down_shexp.weight q8_0 +blk.26.ffn_gate_shexp.weight q8_0 +blk.26.ffn_up_shexp.weight q8_0 +blk.27.attn_kv_a_mqa.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q_a.weight q8_0 +blk.27.attn_q_b.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 +blk.28.attn_kv_a_mqa.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q_a.weight q8_0 +blk.28.attn_q_b.weight q8_0 +blk.28.ffn_down_shexp.weight q8_0 +blk.28.ffn_gate_shexp.weight q8_0 +blk.28.ffn_up_shexp.weight q8_0 +blk.29.attn_kv_a_mqa.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q_a.weight q8_0 +blk.29.attn_q_b.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_gate_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 +blk.30.attn_kv_a_mqa.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q_a.weight q8_0 +blk.30.attn_q_b.weight q8_0 +blk.30.ffn_down_shexp.weight q8_0 +blk.30.ffn_gate_shexp.weight q8_0 +blk.30.ffn_up_shexp.weight q8_0 +blk.31.attn_kv_a_mqa.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q_a.weight q8_0 +blk.31.attn_q_b.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 +blk.32.attn_kv_a_mqa.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q_a.weight q8_0 +blk.32.attn_q_b.weight q8_0 +blk.32.ffn_down_shexp.weight q8_0 +blk.32.ffn_gate_shexp.weight q8_0 +blk.32.ffn_up_shexp.weight q8_0 +blk.33.attn_kv_a_mqa.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q_a.weight q8_0 +blk.33.attn_q_b.weight q8_0 +blk.33.ffn_down_shexp.weight q8_0 +blk.33.ffn_gate_shexp.weight q8_0 +blk.33.ffn_up_shexp.weight q8_0 +blk.34.attn_kv_a_mqa.weight q8_0 +blk.34.attn_output.weight q8_0 +blk.34.attn_q_a.weight q8_0 +blk.34.attn_q_b.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_gate_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 +blk.35.attn_kv_a_mqa.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q_a.weight q8_0 +blk.35.attn_q_b.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 +blk.36.attn_kv_a_mqa.weight q8_0 +blk.36.attn_output.weight q8_0 +blk.36.attn_q_a.weight q8_0 +blk.36.attn_q_b.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_gate_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 +blk.37.attn_kv_a_mqa.weight q8_0 +blk.37.attn_output.weight q8_0 +blk.37.attn_q_a.weight q8_0 +blk.37.attn_q_b.weight q8_0 +blk.37.ffn_down_shexp.weight q8_0 +blk.37.ffn_gate_shexp.weight q8_0 +blk.37.ffn_up_shexp.weight q8_0 +blk.38.attn_kv_a_mqa.weight q8_0 +blk.38.attn_output.weight q8_0 +blk.38.attn_q_a.weight q8_0 +blk.38.attn_q_b.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_gate_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 +blk.39.attn_kv_a_mqa.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q_a.weight q8_0 +blk.39.attn_q_b.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 +blk.40.attn_kv_a_mqa.weight q8_0 +blk.40.attn_output.weight q8_0 +blk.40.attn_q_a.weight q8_0 +blk.40.attn_q_b.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_gate_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 +blk.41.attn_kv_a_mqa.weight q8_0 +blk.41.attn_output.weight q8_0 +blk.41.attn_q_a.weight q8_0 +blk.41.attn_q_b.weight q8_0 +blk.41.ffn_down_shexp.weight q8_0 +blk.41.ffn_gate_shexp.weight q8_0 +blk.41.ffn_up_shexp.weight q8_0 +blk.42.attn_kv_a_mqa.weight q8_0 +blk.42.attn_output.weight q8_0 +blk.42.attn_q_a.weight q8_0 +blk.42.attn_q_b.weight q8_0 +blk.42.ffn_down_shexp.weight q8_0 +blk.42.ffn_gate_shexp.weight q8_0 +blk.42.ffn_up_shexp.weight q8_0 +blk.43.attn_kv_a_mqa.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q_a.weight q8_0 +blk.43.attn_q_b.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 +blk.44.attn_kv_a_mqa.weight q8_0 +blk.44.attn_output.weight q8_0 +blk.44.attn_q_a.weight q8_0 +blk.44.attn_q_b.weight q8_0 +blk.44.ffn_down_shexp.weight q8_0 +blk.44.ffn_gate_shexp.weight q8_0 +blk.44.ffn_up_shexp.weight q8_0 +blk.45.attn_kv_a_mqa.weight q8_0 +blk.45.attn_output.weight q8_0 +blk.45.attn_q_a.weight q8_0 +blk.45.attn_q_b.weight q8_0 +blk.45.ffn_down_shexp.weight q8_0 +blk.45.ffn_gate_shexp.weight q8_0 +blk.45.ffn_up_shexp.weight q8_0 +blk.46.attn_kv_a_mqa.weight q8_0 +blk.46.attn_output.weight q8_0 +blk.46.attn_q_a.weight q8_0 +blk.46.attn_q_b.weight q8_0 +blk.46.ffn_down_shexp.weight q8_0 +blk.46.ffn_gate_shexp.weight q8_0 +blk.46.ffn_up_shexp.weight q8_0 +blk.47.attn_kv_a_mqa.weight q8_0 +blk.47.attn_output.weight q8_0 +blk.47.attn_q_a.weight q8_0 +blk.47.attn_q_b.weight q8_0 +blk.47.ffn_down_shexp.weight q8_0 +blk.47.ffn_gate_shexp.weight q8_0 +blk.47.ffn_up_shexp.weight q8_0 +blk.48.attn_kv_a_mqa.weight q8_0 +blk.48.attn_output.weight q8_0 +blk.48.attn_q_a.weight q8_0 +blk.48.attn_q_b.weight q8_0 +blk.48.ffn_down_shexp.weight q8_0 +blk.48.ffn_gate_shexp.weight q8_0 +blk.48.ffn_up_shexp.weight q8_0 +blk.49.attn_kv_a_mqa.weight q8_0 +blk.49.attn_output.weight q8_0 +blk.49.attn_q_a.weight q8_0 +blk.49.attn_q_b.weight q8_0 +blk.49.ffn_down_shexp.weight q8_0 +blk.49.ffn_gate_shexp.weight q8_0 +blk.49.ffn_up_shexp.weight q8_0 +blk.50.attn_kv_a_mqa.weight q8_0 +blk.50.attn_output.weight q8_0 +blk.50.attn_q_a.weight q8_0 +blk.50.attn_q_b.weight q8_0 +blk.50.ffn_down_shexp.weight q8_0 +blk.50.ffn_gate_shexp.weight q8_0 +blk.50.ffn_up_shexp.weight q8_0 +blk.51.attn_kv_a_mqa.weight q8_0 +blk.51.attn_output.weight q8_0 +blk.51.attn_q_a.weight q8_0 +blk.51.attn_q_b.weight q8_0 +blk.51.ffn_down_shexp.weight q8_0 +blk.51.ffn_gate_shexp.weight q8_0 +blk.51.ffn_up_shexp.weight q8_0 +blk.52.attn_kv_a_mqa.weight q8_0 +blk.52.attn_output.weight q8_0 +blk.52.attn_q_a.weight q8_0 +blk.52.attn_q_b.weight q8_0 +blk.52.ffn_down_shexp.weight q8_0 +blk.52.ffn_gate_shexp.weight q8_0 +blk.52.ffn_up_shexp.weight q8_0 +blk.53.attn_kv_a_mqa.weight q8_0 +blk.53.attn_output.weight q8_0 +blk.53.attn_q_a.weight q8_0 +blk.53.attn_q_b.weight q8_0 +blk.53.ffn_down_shexp.weight q8_0 +blk.53.ffn_gate_shexp.weight q8_0 +blk.53.ffn_up_shexp.weight q8_0 +blk.54.attn_kv_a_mqa.weight q8_0 +blk.54.attn_output.weight q8_0 +blk.54.attn_q_a.weight q8_0 +blk.54.attn_q_b.weight q8_0 +blk.54.ffn_down_shexp.weight q8_0 +blk.54.ffn_gate_shexp.weight q8_0 +blk.54.ffn_up_shexp.weight q8_0 +blk.55.attn_kv_a_mqa.weight q8_0 +blk.55.attn_output.weight q8_0 +blk.55.attn_q_a.weight q8_0 +blk.55.attn_q_b.weight q8_0 +blk.55.ffn_down_shexp.weight q8_0 +blk.55.ffn_gate_shexp.weight q8_0 +blk.55.ffn_up_shexp.weight q8_0 +blk.56.attn_kv_a_mqa.weight q8_0 +blk.56.attn_output.weight q8_0 +blk.56.attn_q_a.weight q8_0 +blk.56.attn_q_b.weight q8_0 +blk.56.ffn_down_shexp.weight q8_0 +blk.56.ffn_gate_shexp.weight q8_0 +blk.56.ffn_up_shexp.weight q8_0 +blk.57.attn_kv_a_mqa.weight q8_0 +blk.57.attn_output.weight q8_0 +blk.57.attn_q_a.weight q8_0 +blk.57.attn_q_b.weight q8_0 +blk.57.ffn_down_shexp.weight q8_0 +blk.57.ffn_gate_shexp.weight q8_0 +blk.57.ffn_up_shexp.weight q8_0 +blk.58.attn_kv_a_mqa.weight q8_0 +blk.58.attn_output.weight q8_0 +blk.58.attn_q_a.weight q8_0 +blk.58.attn_q_b.weight q8_0 +blk.58.ffn_down_shexp.weight q8_0 +blk.58.ffn_gate_shexp.weight q8_0 +blk.58.ffn_up_shexp.weight q8_0 +blk.59.attn_kv_a_mqa.weight q8_0 +blk.59.attn_output.weight q8_0 +blk.59.attn_q_a.weight q8_0 +blk.59.attn_q_b.weight q8_0 +blk.59.ffn_down_shexp.weight q8_0 +blk.59.ffn_gate_shexp.weight q8_0 +blk.59.ffn_up_shexp.weight q8_0 +blk.60.attn_kv_a_mqa.weight q8_0 +blk.60.attn_output.weight q8_0 +blk.60.attn_q_a.weight q8_0 +blk.60.attn_q_b.weight q8_0 +blk.60.ffn_down_shexp.weight q8_0 +blk.60.ffn_gate_shexp.weight q8_0 +blk.60.ffn_up_shexp.weight q8_0 diff --git a/tests/snapshots/gemma-3-4b-it.schema b/tests/snapshots/gemma-3-4b-it.schema new file mode 100644 index 0000000000..1bec284c87 --- /dev/null +++ b/tests/snapshots/gemma-3-4b-it.schema @@ -0,0 +1,1455 @@ +# Model: gemma-3-4b-it +# n_embd=2560, n_ff=10240, n_vocab=262144, n_layer=34, n_head=8, n_head_kv=4 + +[F32] f32 +token_embd.weight q6_K + +[F16] f16 +token_embd.weight q6_K + +[Q4_0] q4_0 +token_embd.weight q6_K + +[Q4_1] q4_1 +token_embd.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +token_embd.weight q6_K + +[Q5_1] q5_1 +token_embd.weight q6_K + +[Q2_K] q2_K +token_embd.weight q6_K +blk.0.ffn_down.weight q3_K +blk.0.attn_output.weight q3_K +blk.0.attn_v.weight q3_K +blk.1.ffn_down.weight q3_K +blk.1.attn_output.weight q3_K +blk.1.attn_v.weight q3_K +blk.10.ffn_down.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q3_K +blk.11.ffn_down.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q3_K +blk.12.ffn_down.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q3_K +blk.13.ffn_down.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q3_K +blk.2.ffn_down.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.attn_v.weight q3_K +blk.3.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q3_K +blk.4.ffn_down.weight q3_K +blk.4.attn_output.weight q3_K +blk.4.attn_v.weight q3_K +blk.5.ffn_down.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q3_K +blk.6.ffn_down.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.attn_v.weight q3_K +blk.7.ffn_down.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q3_K +blk.8.ffn_down.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.attn_v.weight q3_K +blk.9.ffn_down.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.attn_v.weight q3_K +blk.14.ffn_down.weight q3_K +blk.15.ffn_down.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q3_K +blk.16.ffn_down.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.attn_v.weight q3_K +blk.17.ffn_down.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.attn_v.weight q3_K +blk.18.ffn_down.weight q3_K +blk.18.attn_output.weight q3_K +blk.18.attn_v.weight q3_K +blk.19.ffn_down.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q3_K +blk.20.ffn_down.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.attn_v.weight q3_K +blk.21.ffn_down.weight q3_K +blk.21.attn_output.weight q3_K +blk.21.attn_v.weight q3_K +blk.22.ffn_down.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.attn_v.weight q3_K +blk.23.ffn_down.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q3_K +blk.24.ffn_down.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.attn_v.weight q3_K +blk.25.ffn_down.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.attn_v.weight q3_K +blk.26.ffn_down.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q3_K +blk.27.ffn_down.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q3_K +blk.28.ffn_down.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.attn_v.weight q3_K +blk.29.ffn_down.weight q3_K +blk.29.attn_output.weight q3_K +blk.29.attn_v.weight q3_K +blk.30.ffn_down.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.attn_v.weight q3_K +blk.31.ffn_down.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q3_K +blk.32.ffn_down.weight q3_K +blk.32.attn_output.weight q3_K +blk.32.attn_v.weight q3_K +blk.33.ffn_down.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q3_K + +[Q3_K_S] q3_K +token_embd.weight q6_K + +[Q3_K_M] q3_K +token_embd.weight q6_K +blk.0.ffn_down.weight q5_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.10.ffn_down.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.ffn_down.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.ffn_down.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.4.ffn_down.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.ffn_down.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.ffn_down.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.14.ffn_down.weight q4_K +blk.15.ffn_down.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.ffn_down.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.ffn_down.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.ffn_down.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.ffn_down.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.ffn_down.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.ffn_down.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.ffn_down.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.ffn_down.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.ffn_down.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.ffn_down.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.ffn_down.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.ffn_down.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.ffn_down.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.ffn_down.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.ffn_down.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.ffn_down.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.ffn_down.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.ffn_down.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K + +[Q3_K_L] q3_K +token_embd.weight q6_K +blk.0.ffn_down.weight q5_K +blk.0.attn_output.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.attn_v.weight q5_K +blk.10.ffn_down.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.ffn_down.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.ffn_down.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.5.ffn_down.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.ffn_down.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.14.ffn_down.weight q5_K +blk.15.ffn_down.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.ffn_down.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.ffn_down.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.ffn_down.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.ffn_down.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.ffn_down.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.ffn_down.weight q5_K +blk.21.attn_output.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.ffn_down.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.ffn_down.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.ffn_down.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.ffn_down.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.ffn_down.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.ffn_down.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.ffn_down.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.ffn_down.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.ffn_down.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.ffn_down.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.ffn_down.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K + +[Q4_K_S] q4_K +token_embd.weight q6_K +blk.0.ffn_down.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.10.ffn_down.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.11.attn_v.weight q5_K + +[Q4_K_M] q4_K +token_embd.weight q6_K +blk.0.ffn_down.weight q6_K +blk.0.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.10.ffn_down.weight q6_K +blk.10.attn_v.weight q6_K +blk.11.ffn_down.weight q6_K +blk.11.attn_v.weight q6_K +blk.14.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.4.attn_v.weight q6_K +blk.5.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.8.ffn_down.weight q6_K +blk.15.ffn_down.weight q6_K +blk.15.attn_v.weight q6_K +blk.18.ffn_down.weight q6_K +blk.18.attn_v.weight q6_K +blk.21.ffn_down.weight q6_K +blk.21.attn_v.weight q6_K +blk.24.ffn_down.weight q6_K +blk.24.attn_v.weight q6_K +blk.27.ffn_down.weight q6_K +blk.27.attn_v.weight q6_K +blk.29.ffn_down.weight q6_K +blk.29.attn_v.weight q6_K +blk.30.ffn_down.weight q6_K +blk.30.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.31.attn_v.weight q6_K +blk.32.ffn_down.weight q6_K +blk.32.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K +blk.33.attn_v.weight q6_K + +[Q5_K_S] q5_K +token_embd.weight q6_K + +[Q5_K_M] q5_K +token_embd.weight q6_K +blk.0.ffn_down.weight q6_K +blk.0.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.10.ffn_down.weight q6_K +blk.10.attn_v.weight q6_K +blk.11.ffn_down.weight q6_K +blk.11.attn_v.weight q6_K +blk.14.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.4.attn_v.weight q6_K +blk.5.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.8.ffn_down.weight q6_K +blk.15.ffn_down.weight q6_K +blk.15.attn_v.weight q6_K +blk.18.ffn_down.weight q6_K +blk.18.attn_v.weight q6_K +blk.21.ffn_down.weight q6_K +blk.21.attn_v.weight q6_K +blk.24.ffn_down.weight q6_K +blk.24.attn_v.weight q6_K +blk.27.ffn_down.weight q6_K +blk.27.attn_v.weight q6_K +blk.29.ffn_down.weight q6_K +blk.29.attn_v.weight q6_K +blk.30.ffn_down.weight q6_K +blk.30.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.31.attn_v.weight q6_K +blk.32.ffn_down.weight q6_K +blk.32.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K +blk.33.attn_v.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +token_embd.weight q5_K +blk.0.ffn_down.weight q2_K +blk.0.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K +blk.1.attn_v.weight q2_K +blk.10.ffn_down.weight q2_K +blk.10.attn_v.weight q2_K +blk.11.ffn_down.weight q2_K +blk.11.attn_v.weight q2_K +blk.12.attn_v.weight q2_K +blk.13.attn_v.weight q2_K +blk.14.attn_v.weight q2_K +blk.2.attn_v.weight q2_K +blk.3.attn_v.weight q2_K +blk.4.attn_v.weight q2_K +blk.5.attn_v.weight q2_K +blk.6.attn_v.weight q2_K +blk.7.attn_v.weight q2_K +blk.8.attn_v.weight q2_K +blk.9.attn_v.weight q2_K +blk.15.attn_v.weight q2_K +blk.16.attn_v.weight q2_K +blk.17.attn_v.weight q2_K +blk.18.attn_v.weight q2_K +blk.19.attn_v.weight q2_K +blk.20.attn_v.weight q2_K +blk.21.attn_v.weight q2_K +blk.22.attn_v.weight q2_K +blk.23.attn_v.weight q2_K +blk.24.attn_v.weight q2_K +blk.25.attn_v.weight q2_K +blk.26.attn_v.weight q2_K +blk.27.attn_v.weight q2_K +blk.28.attn_v.weight q2_K +blk.29.attn_v.weight q2_K +blk.30.attn_v.weight q2_K +blk.31.attn_v.weight q2_K +blk.32.attn_v.weight q2_K +blk.33.attn_v.weight q2_K + +[IQ2_XS] iq2_xs +token_embd.weight q5_K +blk.0.ffn_down.weight q2_K +blk.0.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K +blk.1.attn_v.weight q2_K +blk.10.ffn_down.weight q2_K +blk.10.attn_v.weight q2_K +blk.11.ffn_down.weight q2_K +blk.11.attn_v.weight q2_K +blk.12.attn_v.weight q2_K +blk.13.attn_v.weight q2_K +blk.14.attn_v.weight q2_K +blk.2.attn_v.weight q2_K +blk.3.attn_v.weight q2_K +blk.4.attn_v.weight q2_K +blk.5.attn_v.weight q2_K +blk.6.attn_v.weight q2_K +blk.7.attn_v.weight q2_K +blk.8.attn_v.weight q2_K +blk.9.attn_v.weight q2_K +blk.15.attn_v.weight q2_K +blk.16.attn_v.weight q2_K +blk.17.attn_v.weight q2_K +blk.18.attn_v.weight q2_K +blk.19.attn_v.weight q2_K +blk.20.attn_v.weight q2_K +blk.21.attn_v.weight q2_K +blk.22.attn_v.weight q2_K +blk.23.attn_v.weight q2_K +blk.24.attn_v.weight q2_K +blk.25.attn_v.weight q2_K +blk.26.attn_v.weight q2_K +blk.27.attn_v.weight q2_K +blk.28.attn_v.weight q2_K +blk.29.attn_v.weight q2_K +blk.30.attn_v.weight q2_K +blk.31.attn_v.weight q2_K +blk.32.attn_v.weight q2_K +blk.33.attn_v.weight q2_K + +[Q2_K_S] q2_K +token_embd.weight q6_K +blk.0.ffn_down.weight q4_K +blk.1.ffn_down.weight q4_K +blk.10.ffn_down.weight q4_K +blk.11.ffn_down.weight q4_K + +[IQ3_XS] iq3_s +token_embd.weight q6_K +blk.0.attn_k.weight iq3_xxs +blk.0.attn_q.weight iq3_xxs +blk.1.attn_k.weight iq3_xxs +blk.1.attn_q.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.2.ffn_gate.weight iq3_xxs +blk.2.ffn_up.weight iq3_xxs +blk.2.attn_k.weight iq3_xxs +blk.2.attn_q.weight iq3_xxs +blk.3.ffn_gate.weight iq3_xxs +blk.3.ffn_up.weight iq3_xxs +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.4.ffn_gate.weight iq3_xxs +blk.4.ffn_up.weight iq3_xxs +blk.4.attn_k.weight iq3_xxs +blk.4.attn_q.weight iq3_xxs +blk.5.ffn_gate.weight iq3_xxs +blk.5.ffn_up.weight iq3_xxs +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.6.ffn_gate.weight iq3_xxs +blk.6.ffn_up.weight iq3_xxs +blk.6.attn_k.weight iq3_xxs +blk.6.attn_q.weight iq3_xxs +blk.7.ffn_gate.weight iq3_xxs +blk.7.ffn_up.weight iq3_xxs +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.8.ffn_gate.weight iq3_xxs +blk.8.ffn_up.weight iq3_xxs +blk.8.attn_k.weight iq3_xxs +blk.8.attn_q.weight iq3_xxs +blk.9.ffn_gate.weight iq3_xxs +blk.9.ffn_up.weight iq3_xxs +blk.9.attn_k.weight iq3_xxs +blk.9.attn_q.weight iq3_xxs +blk.15.ffn_gate.weight iq3_xxs +blk.15.ffn_up.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs +blk.16.attn_k.weight iq3_xxs +blk.16.attn_q.weight iq3_xxs +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs +blk.17.attn_k.weight iq3_xxs +blk.17.attn_q.weight iq3_xxs +blk.18.ffn_gate.weight iq3_xxs +blk.18.ffn_up.weight iq3_xxs +blk.18.attn_k.weight iq3_xxs +blk.18.attn_q.weight iq3_xxs +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs +blk.20.attn_k.weight iq3_xxs +blk.20.attn_q.weight iq3_xxs +blk.21.ffn_gate.weight iq3_xxs +blk.21.ffn_up.weight iq3_xxs +blk.21.attn_k.weight iq3_xxs +blk.21.attn_q.weight iq3_xxs +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs +blk.22.attn_k.weight iq3_xxs +blk.22.attn_q.weight iq3_xxs +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs +blk.24.attn_k.weight iq3_xxs +blk.24.attn_q.weight iq3_xxs +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs +blk.25.attn_k.weight iq3_xxs +blk.25.attn_q.weight iq3_xxs +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs +blk.28.attn_k.weight iq3_xxs +blk.28.attn_q.weight iq3_xxs +blk.29.attn_k.weight iq3_xxs +blk.29.attn_q.weight iq3_xxs +blk.30.attn_k.weight iq3_xxs +blk.30.attn_q.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.32.attn_k.weight iq3_xxs +blk.32.attn_q.weight iq3_xxs +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs + +[IQ3_XXS] iq3_xxs +token_embd.weight q5_K +blk.0.ffn_down.weight q4_K +blk.0.attn_k.weight iq2_s +blk.0.attn_output.weight iq3_s +blk.0.attn_q.weight iq2_s +blk.0.attn_v.weight iq3_s +blk.1.ffn_down.weight q4_K +blk.1.attn_k.weight iq2_s +blk.1.attn_output.weight iq3_s +blk.1.attn_q.weight iq2_s +blk.1.attn_v.weight iq3_s +blk.10.ffn_down.weight q4_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight iq3_s +blk.11.ffn_down.weight q4_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight iq3_s +blk.12.ffn_down.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight iq3_s +blk.13.ffn_down.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight iq3_s +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight iq3_s +blk.2.ffn_down.weight q3_K +blk.2.attn_k.weight iq2_s +blk.2.attn_output.weight iq3_s +blk.2.attn_q.weight iq2_s +blk.2.attn_v.weight iq3_s +blk.3.ffn_down.weight q3_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight iq3_s +blk.4.ffn_down.weight q3_K +blk.4.attn_k.weight iq2_s +blk.4.attn_output.weight iq3_s +blk.4.attn_q.weight iq2_s +blk.4.attn_v.weight iq3_s +blk.5.ffn_down.weight q3_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight iq3_s +blk.6.ffn_down.weight q3_K +blk.6.attn_k.weight iq2_s +blk.6.attn_output.weight iq3_s +blk.6.attn_q.weight iq2_s +blk.6.attn_v.weight iq3_s +blk.7.ffn_down.weight q3_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight iq3_s +blk.8.ffn_down.weight q3_K +blk.8.attn_k.weight iq2_s +blk.8.attn_output.weight iq3_s +blk.8.attn_q.weight iq2_s +blk.8.attn_v.weight iq3_s +blk.9.ffn_down.weight q3_K +blk.9.attn_k.weight iq2_s +blk.9.attn_output.weight iq3_s +blk.9.attn_q.weight iq2_s +blk.9.attn_v.weight iq3_s +blk.14.ffn_down.weight q3_K +blk.15.ffn_down.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight iq3_s +blk.16.ffn_down.weight q3_K +blk.16.attn_k.weight iq2_s +blk.16.attn_output.weight iq3_s +blk.16.attn_q.weight iq2_s +blk.16.attn_v.weight iq3_s +blk.17.ffn_down.weight q3_K +blk.17.attn_k.weight iq2_s +blk.17.attn_output.weight iq3_s +blk.17.attn_q.weight iq2_s +blk.17.attn_v.weight iq3_s +blk.18.ffn_down.weight q3_K +blk.18.attn_k.weight iq2_s +blk.18.attn_output.weight iq3_s +blk.18.attn_q.weight iq2_s +blk.18.attn_v.weight iq3_s +blk.19.ffn_down.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight iq3_s +blk.20.ffn_down.weight q3_K +blk.20.attn_k.weight iq2_s +blk.20.attn_output.weight iq3_s +blk.20.attn_q.weight iq2_s +blk.20.attn_v.weight iq3_s +blk.21.ffn_down.weight q3_K +blk.21.attn_k.weight iq2_s +blk.21.attn_output.weight iq3_s +blk.21.attn_q.weight iq2_s +blk.21.attn_v.weight iq3_s +blk.22.ffn_down.weight q3_K +blk.22.attn_k.weight iq2_s +blk.22.attn_output.weight iq3_s +blk.22.attn_q.weight iq2_s +blk.22.attn_v.weight iq3_s +blk.23.ffn_down.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight iq3_s +blk.24.ffn_down.weight q3_K +blk.24.attn_k.weight iq2_s +blk.24.attn_output.weight iq3_s +blk.24.attn_q.weight iq2_s +blk.24.attn_v.weight iq3_s +blk.25.ffn_down.weight q3_K +blk.25.attn_k.weight iq2_s +blk.25.attn_output.weight iq3_s +blk.25.attn_q.weight iq2_s +blk.25.attn_v.weight iq3_s +blk.26.ffn_down.weight q3_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight iq3_s +blk.27.ffn_down.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight iq3_s +blk.28.ffn_down.weight q3_K +blk.28.attn_k.weight iq2_s +blk.28.attn_output.weight iq3_s +blk.28.attn_q.weight iq2_s +blk.28.attn_v.weight iq3_s +blk.29.ffn_down.weight q3_K +blk.29.attn_k.weight iq2_s +blk.29.attn_output.weight iq3_s +blk.29.attn_q.weight iq2_s +blk.29.attn_v.weight iq3_s +blk.30.ffn_down.weight q3_K +blk.30.attn_k.weight iq2_s +blk.30.attn_output.weight iq3_s +blk.30.attn_q.weight iq2_s +blk.30.attn_v.weight iq3_s +blk.31.ffn_down.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight iq3_s +blk.32.ffn_down.weight q3_K +blk.32.attn_k.weight iq2_s +blk.32.attn_output.weight iq3_s +blk.32.attn_q.weight iq2_s +blk.32.attn_v.weight iq3_s +blk.33.ffn_down.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight iq3_s + +[IQ1_S] iq1_s +token_embd.weight q5_K +blk.0.ffn_down.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q2_K +blk.10.ffn_down.weight q2_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q2_K +blk.11.ffn_down.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q2_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q2_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q2_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q2_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q2_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q2_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q2_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q2_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q2_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q2_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q2_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q2_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q2_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q2_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q2_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q2_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q2_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q2_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q2_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q2_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q2_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q2_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q2_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q2_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q2_K + +[IQ4_NL] iq4_nl +token_embd.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.10.ffn_down.weight q5_K +blk.11.ffn_down.weight q5_K + +[IQ3_S] iq3_s +token_embd.weight q6_K + +[IQ3_M] iq3_s +token_embd.weight q6_K +blk.0.ffn_down.weight q4_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q4_K +blk.10.ffn_down.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K + +[IQ2_S] iq2_xs +token_embd.weight q5_K +blk.0.ffn_down.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight iq3_s +blk.10.ffn_down.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight iq3_s +blk.11.ffn_down.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight iq3_s +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight iq3_s +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight iq3_s +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight iq3_s +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight iq3_s +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight iq3_s +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight iq3_s +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight iq3_s +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight iq3_s +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight iq3_s +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight iq3_s +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight iq3_s +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight iq3_s +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight iq3_s +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight iq3_s +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight iq3_s +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight iq3_s +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight iq3_s +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight iq3_s + +[IQ2_M] iq2_s +token_embd.weight q5_K +blk.0.ffn_down.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight iq3_s +blk.10.ffn_down.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight iq3_s +blk.11.ffn_down.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight iq3_s +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight iq3_s +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight iq3_s +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight iq3_s +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight iq3_s +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight iq3_s +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight iq3_s +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight iq3_s +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight iq3_s +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight iq3_s +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight iq3_s +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight iq3_s +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight iq3_s +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight iq3_s +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight iq3_s +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight iq3_s +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight iq3_s +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight iq3_s +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight iq3_s + +[IQ4_XS] iq4_xs +token_embd.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.10.ffn_down.weight q5_K +blk.11.ffn_down.weight q5_K + +[IQ1_M] iq1_m +token_embd.weight q5_K +blk.0.ffn_down.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q2_K +blk.10.ffn_down.weight q2_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q2_K +blk.11.ffn_down.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q2_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q2_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q2_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q2_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q2_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q2_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q2_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q2_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q2_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q2_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q2_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q2_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q2_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q2_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q2_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q2_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q2_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q2_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q2_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q2_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q2_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q2_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q2_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q2_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q2_K + +[BF16] bf16 +token_embd.weight q6_K + +[TQ1_0] tq1_0 +token_embd.weight q6_K + +[TQ2_0] tq2_0 +token_embd.weight q6_K + +[MXFP4_MOE] mxfp4 +token_embd.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.0.attn_k.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q.weight q8_0 +blk.0.attn_v.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.1.attn_k.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q.weight q8_0 +blk.1.attn_v.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.12.ffn_down.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.2.attn_k.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q.weight q8_0 +blk.2.attn_v.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.4.ffn_down.weight q8_0 +blk.4.ffn_gate.weight q8_0 +blk.4.ffn_up.weight q8_0 +blk.4.attn_k.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q.weight q8_0 +blk.4.attn_v.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 +blk.6.attn_k.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q.weight q8_0 +blk.6.attn_v.weight q8_0 +blk.7.ffn_down.weight q8_0 +blk.7.ffn_gate.weight q8_0 +blk.7.ffn_up.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 +blk.8.attn_k.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q.weight q8_0 +blk.8.attn_v.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 +blk.9.attn_k.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q.weight q8_0 +blk.9.attn_v.weight q8_0 +blk.14.ffn_down.weight q8_0 +blk.15.ffn_down.weight q8_0 +blk.15.ffn_gate.weight q8_0 +blk.15.ffn_up.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 +blk.16.attn_k.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q.weight q8_0 +blk.16.attn_v.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 +blk.17.attn_k.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q.weight q8_0 +blk.17.attn_v.weight q8_0 +blk.18.ffn_down.weight q8_0 +blk.18.ffn_gate.weight q8_0 +blk.18.ffn_up.weight q8_0 +blk.18.attn_k.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q.weight q8_0 +blk.18.attn_v.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 +blk.20.attn_k.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q.weight q8_0 +blk.20.attn_v.weight q8_0 +blk.21.ffn_down.weight q8_0 +blk.21.ffn_gate.weight q8_0 +blk.21.ffn_up.weight q8_0 +blk.21.attn_k.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.21.attn_q.weight q8_0 +blk.21.attn_v.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 +blk.22.attn_k.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q.weight q8_0 +blk.22.attn_v.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 +blk.24.attn_k.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q.weight q8_0 +blk.24.attn_v.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 +blk.25.attn_k.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q.weight q8_0 +blk.25.attn_v.weight q8_0 +blk.26.ffn_down.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 +blk.28.attn_k.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q.weight q8_0 +blk.28.attn_v.weight q8_0 +blk.29.ffn_down.weight q8_0 +blk.29.ffn_gate.weight q8_0 +blk.29.ffn_up.weight q8_0 +blk.29.attn_k.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q.weight q8_0 +blk.29.attn_v.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 +blk.30.attn_k.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q.weight q8_0 +blk.30.attn_v.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.32.ffn_down.weight q8_0 +blk.32.ffn_gate.weight q8_0 +blk.32.ffn_up.weight q8_0 +blk.32.attn_k.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q.weight q8_0 +blk.32.attn_v.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 diff --git a/tests/snapshots/gpt-oss-120b.schema b/tests/snapshots/gpt-oss-120b.schema new file mode 100644 index 0000000000..e66fbcfd2b --- /dev/null +++ b/tests/snapshots/gpt-oss-120b.schema @@ -0,0 +1,1616 @@ +# Model: gpt-oss-120b +# n_embd=2880, n_ff=2880, n_vocab=0, n_layer=36, n_head=64, n_head_kv=8, n_expert=128 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q8_0 +blk.0.attn_output.weight q3_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q3_K +blk.1.attn_output.weight q3_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q3_K +blk.4.attn_output.weight q3_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q3_K +blk.18.attn_output.weight q3_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q3_K +blk.21.attn_output.weight q3_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q3_K +blk.29.attn_output.weight q3_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.32.attn_output.weight q3_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q3_K +blk.34.attn_output.weight q3_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K + +[Q3_K_S] q3_K +output.weight q8_0 + +[Q3_K_M] q3_K +output.weight q8_0 +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down_exps.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q4_K + +[Q3_K_L] q3_K +output.weight q8_0 +blk.0.attn_output.weight q5_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down_exps.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.8.ffn_down_exps.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.9.ffn_down_exps.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.10.ffn_down_exps.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down_exps.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.13.ffn_down_exps.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.14.ffn_down_exps.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.16.ffn_down_exps.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.17.ffn_down_exps.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down_exps.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.20.ffn_down_exps.weight q5_K +blk.21.attn_output.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down_exps.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.22.ffn_down_exps.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.24.ffn_down_exps.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.25.ffn_down_exps.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down_exps.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.28.ffn_down_exps.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down_exps.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.30.ffn_down_exps.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down_exps.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.33.ffn_down_exps.weight q5_K +blk.34.attn_output.weight q5_K +blk.34.attn_v.weight q5_K +blk.34.ffn_down_exps.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight q5_K + +[Q4_K_S] q4_K +output.weight q8_0 +blk.0.attn_v.weight q5_K +blk.0.ffn_down_exps.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K + +[Q4_K_M] q4_K +output.weight q8_0 +blk.0.attn_v.weight q6_K +blk.0.ffn_down_exps.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down_exps.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down_exps.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.6.attn_v.weight q6_K +blk.6.ffn_down_exps.weight q6_K +blk.9.attn_v.weight q6_K +blk.9.ffn_down_exps.weight q6_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down_exps.weight q6_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.18.attn_v.weight q6_K +blk.18.ffn_down_exps.weight q6_K +blk.21.attn_v.weight q6_K +blk.21.ffn_down_exps.weight q6_K +blk.24.attn_v.weight q6_K +blk.24.ffn_down_exps.weight q6_K +blk.27.attn_v.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.30.attn_v.weight q6_K +blk.30.ffn_down_exps.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down_exps.weight q6_K +blk.32.attn_v.weight q6_K +blk.32.ffn_down_exps.weight q6_K +blk.33.attn_v.weight q6_K +blk.33.ffn_down_exps.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down_exps.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down_exps.weight q6_K + +[Q5_K_S] q5_K +output.weight q8_0 + +[Q5_K_M] q5_K +output.weight q8_0 +blk.0.attn_v.weight q6_K +blk.0.ffn_down_exps.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down_exps.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down_exps.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.6.attn_v.weight q6_K +blk.6.ffn_down_exps.weight q6_K +blk.9.attn_v.weight q6_K +blk.9.ffn_down_exps.weight q6_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down_exps.weight q6_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.18.attn_v.weight q6_K +blk.18.ffn_down_exps.weight q6_K +blk.21.attn_v.weight q6_K +blk.21.ffn_down_exps.weight q6_K +blk.24.attn_v.weight q6_K +blk.24.ffn_down_exps.weight q6_K +blk.27.attn_v.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.30.attn_v.weight q6_K +blk.30.ffn_down_exps.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down_exps.weight q6_K +blk.32.attn_v.weight q6_K +blk.32.ffn_down_exps.weight q6_K +blk.33.attn_v.weight q6_K +blk.33.ffn_down_exps.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down_exps.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down_exps.weight q6_K + +[Q6_K] q6_K +output.weight q8_0 + +[IQ2_XXS] iq2_xxs +output.weight q8_0 +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q8_0 +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q8_0 +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K + +[IQ3_XS] iq3_s +output.weight q8_0 +blk.0.attn_k.weight iq3_xxs +blk.0.attn_q.weight iq3_xxs +blk.0.attn_v.weight q4_K +blk.1.attn_k.weight iq3_xxs +blk.1.attn_q.weight iq3_xxs +blk.1.attn_v.weight q4_K +blk.2.attn_k.weight iq3_xxs +blk.2.attn_q.weight iq3_xxs +blk.2.attn_v.weight q4_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.4.attn_k.weight iq3_xxs +blk.4.attn_q.weight iq3_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_gate_exps.weight iq3_xxs +blk.4.ffn_up_exps.weight iq3_xxs +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.5.attn_v.weight q4_K +blk.5.ffn_gate_exps.weight iq3_xxs +blk.5.ffn_up_exps.weight iq3_xxs +blk.6.attn_k.weight iq3_xxs +blk.6.attn_q.weight iq3_xxs +blk.6.attn_v.weight q4_K +blk.6.ffn_gate_exps.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.8.attn_k.weight iq3_xxs +blk.8.attn_q.weight iq3_xxs +blk.8.attn_v.weight q4_K +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.9.attn_k.weight iq3_xxs +blk.9.attn_q.weight iq3_xxs +blk.9.attn_v.weight q4_K +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.10.attn_v.weight q4_K +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.12.attn_v.weight q4_K +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.13.attn_v.weight q4_K +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.14.attn_v.weight q4_K +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.16.attn_k.weight iq3_xxs +blk.16.attn_q.weight iq3_xxs +blk.16.attn_v.weight q4_K +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.17.attn_k.weight iq3_xxs +blk.17.attn_q.weight iq3_xxs +blk.17.attn_v.weight q4_K +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.18.attn_k.weight iq3_xxs +blk.18.attn_q.weight iq3_xxs +blk.18.attn_v.weight q4_K +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.20.attn_k.weight iq3_xxs +blk.20.attn_q.weight iq3_xxs +blk.20.attn_v.weight q4_K +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.21.attn_k.weight iq3_xxs +blk.21.attn_q.weight iq3_xxs +blk.21.attn_v.weight q4_K +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.22.attn_k.weight iq3_xxs +blk.22.attn_q.weight iq3_xxs +blk.22.attn_v.weight q4_K +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.24.attn_k.weight iq3_xxs +blk.24.attn_q.weight iq3_xxs +blk.24.attn_v.weight q4_K +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.25.attn_k.weight iq3_xxs +blk.25.attn_q.weight iq3_xxs +blk.25.attn_v.weight q4_K +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.26.attn_v.weight q4_K +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.28.attn_k.weight iq3_xxs +blk.28.attn_q.weight iq3_xxs +blk.28.attn_v.weight q4_K +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.29.attn_k.weight iq3_xxs +blk.29.attn_q.weight iq3_xxs +blk.29.attn_v.weight q4_K +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.30.attn_k.weight iq3_xxs +blk.30.attn_q.weight iq3_xxs +blk.30.attn_v.weight q4_K +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_k.weight iq3_xxs +blk.32.attn_q.weight iq3_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_k.weight iq3_xxs +blk.34.attn_q.weight iq3_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K + +[IQ3_XXS] iq3_xxs +output.weight q8_0 +token_embd.weight iq3_s +blk.0.attn_k.weight iq2_s +blk.0.attn_output.weight iq3_s +blk.0.attn_q.weight iq2_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q4_K +blk.1.attn_k.weight iq2_s +blk.1.attn_output.weight iq3_s +blk.1.attn_q.weight iq2_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.attn_k.weight iq2_s +blk.2.attn_output.weight iq3_s +blk.2.attn_q.weight iq2_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_k.weight iq2_s +blk.4.attn_output.weight iq3_s +blk.4.attn_q.weight iq2_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q3_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q3_K +blk.6.attn_k.weight iq2_s +blk.6.attn_output.weight iq3_s +blk.6.attn_q.weight iq2_s +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q3_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.8.attn_k.weight iq2_s +blk.8.attn_output.weight iq3_s +blk.8.attn_q.weight iq2_s +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q3_K +blk.9.attn_k.weight iq2_s +blk.9.attn_output.weight iq3_s +blk.9.attn_q.weight iq2_s +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q3_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q3_K +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.16.attn_k.weight iq2_s +blk.16.attn_output.weight iq3_s +blk.16.attn_q.weight iq2_s +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q3_K +blk.17.attn_k.weight iq2_s +blk.17.attn_output.weight iq3_s +blk.17.attn_q.weight iq2_s +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q3_K +blk.18.attn_k.weight iq2_s +blk.18.attn_output.weight iq3_s +blk.18.attn_q.weight iq2_s +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.20.attn_k.weight iq2_s +blk.20.attn_output.weight iq3_s +blk.20.attn_q.weight iq2_s +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q3_K +blk.21.attn_k.weight iq2_s +blk.21.attn_output.weight iq3_s +blk.21.attn_q.weight iq2_s +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q3_K +blk.22.attn_k.weight iq2_s +blk.22.attn_output.weight iq3_s +blk.22.attn_q.weight iq2_s +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.24.attn_k.weight iq2_s +blk.24.attn_output.weight iq3_s +blk.24.attn_q.weight iq2_s +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q3_K +blk.25.attn_k.weight iq2_s +blk.25.attn_output.weight iq3_s +blk.25.attn_q.weight iq2_s +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q3_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.28.attn_k.weight iq2_s +blk.28.attn_output.weight iq3_s +blk.28.attn_q.weight iq2_s +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q3_K +blk.29.attn_k.weight iq2_s +blk.29.attn_output.weight iq3_s +blk.29.attn_q.weight iq2_s +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q3_K +blk.30.attn_k.weight iq2_s +blk.30.attn_output.weight iq3_s +blk.30.attn_q.weight iq2_s +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.32.attn_k.weight iq2_s +blk.32.attn_output.weight iq3_s +blk.32.attn_q.weight iq2_s +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q3_K +blk.34.attn_k.weight iq2_s +blk.34.attn_output.weight iq3_s +blk.34.attn_q.weight iq2_s +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K + +[IQ1_S] iq1_s +output.weight q8_0 +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down_exps.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.attn_v.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K + +[IQ3_S] iq3_s +output.weight q8_0 +blk.0.attn_v.weight q4_K +blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K + +[IQ3_M] iq3_s +output.weight q8_0 +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q4_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q8_0 +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q8_0 +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q8_0 +blk.0.attn_v.weight q5_K +blk.0.ffn_down_exps.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.attn_v.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K + +[IQ1_M] iq1_m +output.weight q8_0 +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down_exps.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q8_0 +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q8_0 +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_k.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q.weight q8_0 +blk.0.attn_v.weight q8_0 +blk.1.attn_k.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q.weight q8_0 +blk.1.attn_v.weight q8_0 +blk.2.attn_k.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q.weight q8_0 +blk.2.attn_v.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.4.attn_k.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q.weight q8_0 +blk.4.attn_v.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.6.attn_k.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q.weight q8_0 +blk.6.attn_v.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.8.attn_k.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q.weight q8_0 +blk.8.attn_v.weight q8_0 +blk.9.attn_k.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q.weight q8_0 +blk.9.attn_v.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.16.attn_k.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q.weight q8_0 +blk.16.attn_v.weight q8_0 +blk.17.attn_k.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q.weight q8_0 +blk.17.attn_v.weight q8_0 +blk.18.attn_k.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q.weight q8_0 +blk.18.attn_v.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.20.attn_k.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q.weight q8_0 +blk.20.attn_v.weight q8_0 +blk.21.attn_k.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.21.attn_q.weight q8_0 +blk.21.attn_v.weight q8_0 +blk.22.attn_k.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q.weight q8_0 +blk.22.attn_v.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.24.attn_k.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q.weight q8_0 +blk.24.attn_v.weight q8_0 +blk.25.attn_k.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q.weight q8_0 +blk.25.attn_v.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.28.attn_k.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q.weight q8_0 +blk.28.attn_v.weight q8_0 +blk.29.attn_k.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q.weight q8_0 +blk.29.attn_v.weight q8_0 +blk.30.attn_k.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q.weight q8_0 +blk.30.attn_v.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.32.attn_k.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q.weight q8_0 +blk.32.attn_v.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 +blk.34.attn_k.weight q8_0 +blk.34.attn_output.weight q8_0 +blk.34.attn_q.weight q8_0 +blk.34.attn_v.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 diff --git a/tests/snapshots/meta-llama-3.1-70b-instruct.schema b/tests/snapshots/meta-llama-3.1-70b-instruct.schema new file mode 100644 index 0000000000..b26755d6f7 --- /dev/null +++ b/tests/snapshots/meta-llama-3.1-70b-instruct.schema @@ -0,0 +1,3899 @@ +# Model: Meta-Llama-3.1-70B-Instruct +# n_embd=8192, n_ff=28672, n_vocab=128256, n_layer=80, n_head=64, n_head_kv=8 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +blk.0.ffn_down.weight q3_K +blk.0.attn_output.weight q3_K +blk.0.attn_v.weight q5_K +blk.1.attn_output.weight q3_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q3_K +blk.2.ffn_down.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q5_K +blk.4.attn_output.weight q3_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q3_K +blk.5.ffn_down.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q3_K +blk.7.attn_output.weight q3_K +blk.8.ffn_down.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.attn_v.weight q5_K +blk.10.ffn_down.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q5_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q5_K +blk.14.ffn_down.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q5_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.attn_v.weight q5_K +blk.17.ffn_down.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.attn_v.weight q5_K +blk.18.attn_output.weight q3_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q5_K +blk.20.ffn_down.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down.weight q3_K +blk.21.attn_output.weight q3_K +blk.22.ffn_down.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.attn_v.weight q5_K +blk.23.ffn_down.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.attn_v.weight q5_K +blk.25.ffn_down.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.attn_v.weight q5_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q5_K +blk.28.ffn_down.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.attn_v.weight q5_K +blk.29.attn_output.weight q3_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.attn_v.weight q5_K +blk.31.ffn_down.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q5_K +blk.32.attn_output.weight q3_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q5_K +blk.34.ffn_down.weight q3_K +blk.34.attn_output.weight q3_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q3_K +blk.35.attn_output.weight q3_K +blk.36.ffn_down.weight q3_K +blk.36.attn_output.weight q3_K +blk.36.attn_v.weight q5_K +blk.37.ffn_down.weight q3_K +blk.37.attn_output.weight q3_K +blk.37.attn_v.weight q5_K +blk.38.ffn_down.weight q3_K +blk.38.attn_output.weight q3_K +blk.38.attn_v.weight q5_K +blk.39.ffn_down.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q5_K +blk.40.attn_output.weight q3_K +blk.40.attn_v.weight q5_K +blk.40.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K +blk.41.attn_output.weight q3_K +blk.41.attn_v.weight q5_K +blk.42.ffn_down.weight q3_K +blk.42.attn_output.weight q3_K +blk.42.attn_v.weight q5_K +blk.43.attn_output.weight q3_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K +blk.44.attn_output.weight q3_K +blk.44.attn_v.weight q5_K +blk.45.ffn_down.weight q3_K +blk.45.attn_output.weight q3_K +blk.45.attn_v.weight q5_K +blk.46.attn_output.weight q3_K +blk.46.attn_v.weight q5_K +blk.46.ffn_down.weight q3_K +blk.47.ffn_down.weight q3_K +blk.47.attn_output.weight q3_K +blk.47.attn_v.weight q5_K +blk.48.ffn_down.weight q3_K +blk.48.attn_output.weight q3_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.49.ffn_down.weight q3_K +blk.49.attn_output.weight q3_K +blk.50.ffn_down.weight q3_K +blk.50.attn_output.weight q3_K +blk.50.attn_v.weight q5_K +blk.51.ffn_down.weight q3_K +blk.51.attn_output.weight q3_K +blk.51.attn_v.weight q5_K +blk.52.ffn_down.weight q3_K +blk.52.attn_output.weight q3_K +blk.52.attn_v.weight q5_K +blk.53.ffn_down.weight q3_K +blk.53.attn_output.weight q3_K +blk.53.attn_v.weight q5_K +blk.54.attn_output.weight q3_K +blk.54.attn_v.weight q5_K +blk.54.ffn_down.weight q3_K +blk.55.ffn_down.weight q3_K +blk.55.attn_output.weight q3_K +blk.55.attn_v.weight q5_K +blk.56.ffn_down.weight q3_K +blk.56.attn_output.weight q3_K +blk.56.attn_v.weight q5_K +blk.57.attn_output.weight q3_K +blk.57.attn_v.weight q5_K +blk.57.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K +blk.58.attn_output.weight q3_K +blk.58.attn_v.weight q5_K +blk.59.ffn_down.weight q3_K +blk.59.attn_output.weight q3_K +blk.59.attn_v.weight q5_K +blk.60.attn_output.weight q3_K +blk.60.attn_v.weight q5_K +blk.60.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K +blk.61.attn_output.weight q3_K +blk.61.attn_v.weight q5_K +blk.62.ffn_down.weight q3_K +blk.62.attn_output.weight q3_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.63.ffn_down.weight q3_K +blk.63.attn_output.weight q3_K +blk.64.ffn_down.weight q3_K +blk.64.attn_output.weight q3_K +blk.64.attn_v.weight q5_K +blk.65.ffn_down.weight q3_K +blk.65.attn_output.weight q3_K +blk.65.attn_v.weight q5_K +blk.66.ffn_down.weight q3_K +blk.66.attn_output.weight q3_K +blk.66.attn_v.weight q5_K +blk.67.ffn_down.weight q3_K +blk.67.attn_output.weight q3_K +blk.67.attn_v.weight q5_K +blk.68.attn_output.weight q3_K +blk.68.attn_v.weight q5_K +blk.68.ffn_down.weight q3_K +blk.69.ffn_down.weight q3_K +blk.69.attn_output.weight q3_K +blk.69.attn_v.weight q5_K +blk.70.ffn_down.weight q3_K +blk.70.attn_output.weight q3_K +blk.70.attn_v.weight q5_K +blk.71.attn_output.weight q3_K +blk.71.attn_v.weight q5_K +blk.71.ffn_down.weight q3_K +blk.72.ffn_down.weight q3_K +blk.72.attn_output.weight q3_K +blk.72.attn_v.weight q5_K +blk.73.ffn_down.weight q3_K +blk.73.attn_output.weight q3_K +blk.73.attn_v.weight q5_K +blk.74.attn_output.weight q3_K +blk.74.attn_v.weight q5_K +blk.74.ffn_down.weight q3_K +blk.75.ffn_down.weight q3_K +blk.75.attn_output.weight q3_K +blk.75.attn_v.weight q5_K +blk.76.ffn_down.weight q3_K +blk.76.attn_output.weight q3_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.77.ffn_down.weight q3_K +blk.77.attn_output.weight q3_K +blk.78.ffn_down.weight q3_K +blk.78.attn_output.weight q3_K +blk.78.attn_v.weight q5_K +blk.79.ffn_down.weight q3_K +blk.79.attn_output.weight q3_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[Q3_K_S] q3_K +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[Q3_K_M] q3_K +blk.0.ffn_down.weight q5_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q5_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.8.ffn_down.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q5_K +blk.10.ffn_down.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q5_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q4_K +blk.13.ffn_down.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q5_K +blk.14.ffn_down.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q5_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q4_K +blk.16.ffn_down.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q5_K +blk.17.ffn_down.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q5_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down.weight q4_K +blk.19.ffn_down.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q5_K +blk.20.ffn_down.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down.weight q4_K +blk.21.attn_output.weight q4_K +blk.22.ffn_down.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q5_K +blk.23.ffn_down.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q5_K +blk.25.ffn_down.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q5_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down.weight q4_K +blk.27.ffn_down.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q5_K +blk.28.ffn_down.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q5_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down.weight q4_K +blk.30.ffn_down.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q5_K +blk.31.ffn_down.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q5_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down.weight q4_K +blk.33.ffn_down.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q5_K +blk.34.ffn_down.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q4_K +blk.35.attn_output.weight q4_K +blk.36.ffn_down.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q5_K +blk.37.ffn_down.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q5_K +blk.38.ffn_down.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q5_K +blk.39.ffn_down.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q5_K +blk.40.attn_output.weight q4_K +blk.40.attn_v.weight q5_K +blk.40.ffn_down.weight q4_K +blk.41.ffn_down.weight q4_K +blk.41.attn_output.weight q4_K +blk.41.attn_v.weight q5_K +blk.42.ffn_down.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q5_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down.weight q4_K +blk.44.ffn_down.weight q4_K +blk.44.attn_output.weight q4_K +blk.44.attn_v.weight q5_K +blk.45.ffn_down.weight q4_K +blk.45.attn_output.weight q4_K +blk.45.attn_v.weight q5_K +blk.46.attn_output.weight q4_K +blk.46.attn_v.weight q5_K +blk.46.ffn_down.weight q4_K +blk.47.ffn_down.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q5_K +blk.48.ffn_down.weight q4_K +blk.48.attn_output.weight q4_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.49.ffn_down.weight q4_K +blk.49.attn_output.weight q4_K +blk.50.ffn_down.weight q4_K +blk.50.attn_output.weight q4_K +blk.50.attn_v.weight q5_K +blk.51.ffn_down.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q5_K +blk.52.ffn_down.weight q4_K +blk.52.attn_output.weight q4_K +blk.52.attn_v.weight q5_K +blk.53.ffn_down.weight q4_K +blk.53.attn_output.weight q4_K +blk.53.attn_v.weight q5_K +blk.54.attn_output.weight q4_K +blk.54.attn_v.weight q5_K +blk.54.ffn_down.weight q4_K +blk.55.ffn_down.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q5_K +blk.56.ffn_down.weight q4_K +blk.56.attn_output.weight q4_K +blk.56.attn_v.weight q5_K +blk.57.attn_output.weight q4_K +blk.57.attn_v.weight q5_K +blk.57.ffn_down.weight q4_K +blk.58.ffn_down.weight q4_K +blk.58.attn_output.weight q4_K +blk.58.attn_v.weight q5_K +blk.59.ffn_down.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q5_K +blk.60.attn_output.weight q4_K +blk.60.attn_v.weight q5_K +blk.60.ffn_down.weight q4_K +blk.61.ffn_down.weight q4_K +blk.61.attn_output.weight q4_K +blk.61.attn_v.weight q5_K +blk.62.ffn_down.weight q4_K +blk.62.attn_output.weight q4_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.63.ffn_down.weight q4_K +blk.63.attn_output.weight q4_K +blk.64.ffn_down.weight q4_K +blk.64.attn_output.weight q4_K +blk.64.attn_v.weight q5_K +blk.65.ffn_down.weight q4_K +blk.65.attn_output.weight q4_K +blk.65.attn_v.weight q5_K +blk.66.ffn_down.weight q4_K +blk.66.attn_output.weight q4_K +blk.66.attn_v.weight q5_K +blk.67.ffn_down.weight q4_K +blk.67.attn_output.weight q4_K +blk.67.attn_v.weight q5_K +blk.68.attn_output.weight q4_K +blk.68.attn_v.weight q5_K +blk.68.ffn_down.weight q4_K +blk.69.ffn_down.weight q4_K +blk.69.attn_output.weight q4_K +blk.69.attn_v.weight q5_K +blk.70.ffn_down.weight q4_K +blk.70.attn_output.weight q4_K +blk.70.attn_v.weight q5_K +blk.71.attn_output.weight q4_K +blk.71.attn_v.weight q5_K +blk.71.ffn_down.weight q4_K +blk.72.ffn_down.weight q4_K +blk.72.attn_output.weight q4_K +blk.72.attn_v.weight q5_K +blk.73.ffn_down.weight q4_K +blk.73.attn_output.weight q4_K +blk.73.attn_v.weight q5_K +blk.74.attn_output.weight q4_K +blk.74.attn_v.weight q5_K +blk.74.ffn_down.weight q4_K +blk.75.ffn_down.weight q4_K +blk.75.attn_output.weight q4_K +blk.75.attn_v.weight q5_K +blk.76.ffn_down.weight q4_K +blk.76.attn_output.weight q4_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.77.ffn_down.weight q4_K +blk.77.attn_output.weight q4_K +blk.78.ffn_down.weight q4_K +blk.78.attn_output.weight q4_K +blk.78.attn_v.weight q5_K +blk.79.ffn_down.weight q4_K +blk.79.attn_output.weight q4_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[Q3_K_L] q3_K +blk.0.ffn_down.weight q5_K +blk.0.attn_output.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.7.attn_output.weight q5_K +blk.8.ffn_down.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.ffn_down.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q5_K +blk.13.ffn_down.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.ffn_down.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q5_K +blk.16.ffn_down.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.ffn_down.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down.weight q5_K +blk.19.ffn_down.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.ffn_down.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down.weight q5_K +blk.21.attn_output.weight q5_K +blk.22.ffn_down.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.ffn_down.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.ffn_down.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down.weight q5_K +blk.27.ffn_down.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.ffn_down.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down.weight q5_K +blk.30.ffn_down.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.ffn_down.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down.weight q5_K +blk.33.ffn_down.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.ffn_down.weight q5_K +blk.34.attn_output.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q5_K +blk.35.attn_output.weight q5_K +blk.36.ffn_down.weight q5_K +blk.36.attn_output.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.ffn_down.weight q5_K +blk.37.attn_output.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.ffn_down.weight q5_K +blk.38.attn_output.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.ffn_down.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_output.weight q5_K +blk.40.attn_v.weight q5_K +blk.40.ffn_down.weight q5_K +blk.41.ffn_down.weight q5_K +blk.41.attn_output.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.ffn_down.weight q5_K +blk.42.attn_output.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down.weight q5_K +blk.44.ffn_down.weight q5_K +blk.44.attn_output.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.ffn_down.weight q5_K +blk.45.attn_output.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_output.weight q5_K +blk.46.attn_v.weight q5_K +blk.46.ffn_down.weight q5_K +blk.47.ffn_down.weight q5_K +blk.47.attn_output.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.ffn_down.weight q5_K +blk.48.attn_output.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.49.ffn_down.weight q5_K +blk.49.attn_output.weight q5_K +blk.50.ffn_down.weight q5_K +blk.50.attn_output.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.ffn_down.weight q5_K +blk.51.attn_output.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.ffn_down.weight q5_K +blk.52.attn_output.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.ffn_down.weight q5_K +blk.53.attn_output.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_output.weight q5_K +blk.54.attn_v.weight q5_K +blk.54.ffn_down.weight q5_K +blk.55.ffn_down.weight q5_K +blk.55.attn_output.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.ffn_down.weight q5_K +blk.56.attn_output.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_output.weight q5_K +blk.57.attn_v.weight q5_K +blk.57.ffn_down.weight q5_K +blk.58.ffn_down.weight q5_K +blk.58.attn_output.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.ffn_down.weight q5_K +blk.59.attn_output.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_output.weight q5_K +blk.60.attn_v.weight q5_K +blk.60.ffn_down.weight q5_K +blk.61.ffn_down.weight q5_K +blk.61.attn_output.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.ffn_down.weight q5_K +blk.62.attn_output.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.63.ffn_down.weight q5_K +blk.63.attn_output.weight q5_K +blk.64.ffn_down.weight q5_K +blk.64.attn_output.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.ffn_down.weight q5_K +blk.65.attn_output.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.ffn_down.weight q5_K +blk.66.attn_output.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.ffn_down.weight q5_K +blk.67.attn_output.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_output.weight q5_K +blk.68.attn_v.weight q5_K +blk.68.ffn_down.weight q5_K +blk.69.ffn_down.weight q5_K +blk.69.attn_output.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.ffn_down.weight q5_K +blk.70.attn_output.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_output.weight q5_K +blk.71.attn_v.weight q5_K +blk.71.ffn_down.weight q5_K +blk.72.ffn_down.weight q5_K +blk.72.attn_output.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.ffn_down.weight q5_K +blk.73.attn_output.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_output.weight q5_K +blk.74.attn_v.weight q5_K +blk.74.ffn_down.weight q5_K +blk.75.ffn_down.weight q5_K +blk.75.attn_output.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.ffn_down.weight q5_K +blk.76.attn_output.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.77.ffn_down.weight q5_K +blk.77.attn_output.weight q5_K +blk.78.ffn_down.weight q5_K +blk.78.attn_output.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.ffn_down.weight q5_K +blk.79.attn_output.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[Q4_K_S] q4_K +blk.0.ffn_down.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[Q4_K_M] q4_K +blk.0.ffn_down.weight q6_K +blk.0.attn_v.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down.weight q6_K +blk.5.ffn_down.weight q6_K +blk.5.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K +blk.6.attn_v.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.8.ffn_down.weight q6_K +blk.8.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K +blk.9.attn_v.weight q6_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down.weight q6_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down.weight q6_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q6_K +blk.18.ffn_down.weight q6_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q6_K +blk.21.ffn_down.weight q6_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q6_K +blk.24.attn_v.weight q6_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.ffn_down.weight q6_K +blk.27.attn_v.weight q6_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.ffn_down.weight q6_K +blk.30.attn_v.weight q6_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.ffn_down.weight q6_K +blk.33.attn_v.weight q6_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.ffn_down.weight q6_K +blk.36.attn_v.weight q6_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.ffn_down.weight q6_K +blk.39.attn_v.weight q6_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.ffn_down.weight q6_K +blk.42.attn_v.weight q6_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.ffn_down.weight q6_K +blk.45.attn_v.weight q6_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.ffn_down.weight q6_K +blk.48.attn_v.weight q6_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.ffn_down.weight q6_K +blk.51.attn_v.weight q6_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q6_K +blk.54.ffn_down.weight q6_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q6_K +blk.57.ffn_down.weight q6_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q6_K +blk.60.ffn_down.weight q6_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q6_K +blk.63.ffn_down.weight q6_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.ffn_down.weight q6_K +blk.66.attn_v.weight q6_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.ffn_down.weight q6_K +blk.69.attn_v.weight q6_K +blk.70.ffn_down.weight q6_K +blk.70.attn_v.weight q6_K +blk.71.attn_v.weight q6_K +blk.71.ffn_down.weight q6_K +blk.72.ffn_down.weight q6_K +blk.72.attn_v.weight q6_K +blk.73.ffn_down.weight q6_K +blk.73.attn_v.weight q6_K +blk.74.attn_v.weight q6_K +blk.74.ffn_down.weight q6_K +blk.75.ffn_down.weight q6_K +blk.75.attn_v.weight q6_K +blk.76.ffn_down.weight q6_K +blk.76.attn_v.weight q6_K +blk.77.attn_v.weight q6_K +blk.77.ffn_down.weight q6_K +blk.78.ffn_down.weight q6_K +blk.78.attn_v.weight q6_K +blk.79.ffn_down.weight q6_K +blk.79.attn_v.weight q6_K +output.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +blk.0.ffn_down.weight q6_K +blk.0.attn_v.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down.weight q6_K +blk.5.ffn_down.weight q6_K +blk.5.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K +blk.6.attn_v.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.8.ffn_down.weight q6_K +blk.8.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K +blk.9.attn_v.weight q6_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down.weight q6_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down.weight q6_K +blk.18.attn_v.weight q6_K +blk.18.ffn_down.weight q6_K +blk.21.attn_v.weight q6_K +blk.21.ffn_down.weight q6_K +blk.24.ffn_down.weight q6_K +blk.24.attn_v.weight q6_K +blk.27.ffn_down.weight q6_K +blk.27.attn_v.weight q6_K +blk.30.ffn_down.weight q6_K +blk.30.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K +blk.33.attn_v.weight q6_K +blk.36.ffn_down.weight q6_K +blk.36.attn_v.weight q6_K +blk.39.ffn_down.weight q6_K +blk.39.attn_v.weight q6_K +blk.42.ffn_down.weight q6_K +blk.42.attn_v.weight q6_K +blk.45.ffn_down.weight q6_K +blk.45.attn_v.weight q6_K +blk.48.ffn_down.weight q6_K +blk.48.attn_v.weight q6_K +blk.51.ffn_down.weight q6_K +blk.51.attn_v.weight q6_K +blk.54.attn_v.weight q6_K +blk.54.ffn_down.weight q6_K +blk.57.attn_v.weight q6_K +blk.57.ffn_down.weight q6_K +blk.60.attn_v.weight q6_K +blk.60.ffn_down.weight q6_K +blk.63.attn_v.weight q6_K +blk.63.ffn_down.weight q6_K +blk.66.ffn_down.weight q6_K +blk.66.attn_v.weight q6_K +blk.69.ffn_down.weight q6_K +blk.69.attn_v.weight q6_K +blk.70.ffn_down.weight q6_K +blk.70.attn_v.weight q6_K +blk.71.attn_v.weight q6_K +blk.71.ffn_down.weight q6_K +blk.72.ffn_down.weight q6_K +blk.72.attn_v.weight q6_K +blk.73.ffn_down.weight q6_K +blk.73.attn_v.weight q6_K +blk.74.attn_v.weight q6_K +blk.74.ffn_down.weight q6_K +blk.75.ffn_down.weight q6_K +blk.75.attn_v.weight q6_K +blk.76.ffn_down.weight q6_K +blk.76.attn_v.weight q6_K +blk.77.attn_v.weight q6_K +blk.77.ffn_down.weight q6_K +blk.78.ffn_down.weight q6_K +blk.78.attn_v.weight q6_K +blk.79.ffn_down.weight q6_K +blk.79.attn_v.weight q6_K +output.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.0.attn_v.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K +blk.45.attn_v.weight q4_K +blk.46.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.50.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.52.attn_v.weight q4_K +blk.53.attn_v.weight q4_K +blk.54.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.56.attn_v.weight q4_K +blk.57.attn_v.weight q4_K +blk.58.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.60.attn_v.weight q4_K +blk.61.attn_v.weight q4_K +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.64.attn_v.weight q4_K +blk.65.attn_v.weight q4_K +blk.66.attn_v.weight q4_K +blk.67.attn_v.weight q4_K +blk.68.attn_v.weight q4_K +blk.69.attn_v.weight q4_K +blk.70.attn_v.weight q4_K +blk.71.attn_v.weight q4_K +blk.72.attn_v.weight q4_K +blk.73.attn_v.weight q4_K +blk.74.attn_v.weight q4_K +blk.75.attn_v.weight q4_K +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.78.attn_v.weight q4_K +blk.79.attn_v.weight q4_K +output.weight q5_K + +[IQ2_XS] iq2_xs +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.0.attn_v.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K +blk.45.attn_v.weight q4_K +blk.46.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.50.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.52.attn_v.weight q4_K +blk.53.attn_v.weight q4_K +blk.54.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.56.attn_v.weight q4_K +blk.57.attn_v.weight q4_K +blk.58.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.60.attn_v.weight q4_K +blk.61.attn_v.weight q4_K +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.64.attn_v.weight q4_K +blk.65.attn_v.weight q4_K +blk.66.attn_v.weight q4_K +blk.67.attn_v.weight q4_K +blk.68.attn_v.weight q4_K +blk.69.attn_v.weight q4_K +blk.70.attn_v.weight q4_K +blk.71.attn_v.weight q4_K +blk.72.attn_v.weight q4_K +blk.73.attn_v.weight q4_K +blk.74.attn_v.weight q4_K +blk.75.attn_v.weight q4_K +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.78.attn_v.weight q4_K +blk.79.attn_v.weight q4_K +output.weight q5_K + +[Q2_K_S] q2_K +blk.0.ffn_down.weight q4_K +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ3_XS] iq3_s +blk.0.attn_k.weight iq3_xxs +blk.0.attn_q.weight iq3_xxs +blk.0.attn_v.weight q5_K +blk.1.attn_k.weight iq3_xxs +blk.1.attn_q.weight iq3_xxs +blk.1.attn_v.weight q5_K +blk.2.attn_k.weight iq3_xxs +blk.2.attn_q.weight iq3_xxs +blk.2.attn_v.weight q5_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q5_K +blk.4.attn_k.weight iq3_xxs +blk.4.attn_q.weight iq3_xxs +blk.4.attn_v.weight q5_K +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.5.attn_v.weight q5_K +blk.6.attn_k.weight iq3_xxs +blk.6.attn_q.weight iq3_xxs +blk.6.attn_v.weight q5_K +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q5_K +blk.8.attn_k.weight iq3_xxs +blk.8.attn_q.weight iq3_xxs +blk.8.attn_v.weight q5_K +blk.9.attn_k.weight iq3_xxs +blk.9.attn_q.weight iq3_xxs +blk.9.attn_v.weight q5_K +blk.10.ffn_gate.weight iq3_xxs +blk.10.ffn_up.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.10.attn_v.weight q5_K +blk.11.ffn_gate.weight iq3_xxs +blk.11.ffn_up.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q5_K +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.12.attn_v.weight q5_K +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.13.attn_v.weight q5_K +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.14.attn_v.weight q5_K +blk.15.ffn_gate.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q5_K +blk.15.ffn_up.weight iq3_xxs +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs +blk.16.attn_k.weight iq3_xxs +blk.16.attn_q.weight iq3_xxs +blk.16.attn_v.weight q5_K +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs +blk.17.attn_k.weight iq3_xxs +blk.17.attn_q.weight iq3_xxs +blk.17.attn_v.weight q5_K +blk.18.attn_k.weight iq3_xxs +blk.18.attn_q.weight iq3_xxs +blk.18.attn_v.weight q5_K +blk.18.ffn_gate.weight iq3_xxs +blk.18.ffn_up.weight iq3_xxs +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q5_K +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs +blk.20.attn_k.weight iq3_xxs +blk.20.attn_q.weight iq3_xxs +blk.20.attn_v.weight q5_K +blk.21.attn_k.weight iq3_xxs +blk.21.attn_q.weight iq3_xxs +blk.21.attn_v.weight q5_K +blk.21.ffn_gate.weight iq3_xxs +blk.21.ffn_up.weight iq3_xxs +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs +blk.22.attn_k.weight iq3_xxs +blk.22.attn_q.weight iq3_xxs +blk.22.attn_v.weight q5_K +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q5_K +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs +blk.24.attn_k.weight iq3_xxs +blk.24.attn_q.weight iq3_xxs +blk.24.attn_v.weight q5_K +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs +blk.25.attn_k.weight iq3_xxs +blk.25.attn_q.weight iq3_xxs +blk.25.attn_v.weight q5_K +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.26.attn_v.weight q5_K +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q5_K +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs +blk.28.attn_k.weight iq3_xxs +blk.28.attn_q.weight iq3_xxs +blk.28.attn_v.weight q5_K +blk.29.ffn_gate.weight iq3_xxs +blk.29.attn_k.weight iq3_xxs +blk.29.attn_q.weight iq3_xxs +blk.29.attn_v.weight q5_K +blk.29.ffn_up.weight iq3_xxs +blk.30.ffn_gate.weight iq3_xxs +blk.30.ffn_up.weight iq3_xxs +blk.30.attn_k.weight iq3_xxs +blk.30.attn_q.weight iq3_xxs +blk.30.attn_v.weight q5_K +blk.31.ffn_gate.weight iq3_xxs +blk.31.ffn_up.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q5_K +blk.32.attn_k.weight iq3_xxs +blk.32.attn_q.weight iq3_xxs +blk.32.attn_v.weight q5_K +blk.32.ffn_gate.weight iq3_xxs +blk.32.ffn_up.weight iq3_xxs +blk.33.ffn_gate.weight iq3_xxs +blk.33.ffn_up.weight iq3_xxs +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs +blk.33.attn_v.weight q5_K +blk.34.ffn_gate.weight iq3_xxs +blk.34.ffn_up.weight iq3_xxs +blk.34.attn_k.weight iq3_xxs +blk.34.attn_q.weight iq3_xxs +blk.34.attn_v.weight q5_K +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q5_K +blk.35.ffn_gate.weight iq3_xxs +blk.35.ffn_up.weight iq3_xxs +blk.36.ffn_gate.weight iq3_xxs +blk.36.ffn_up.weight iq3_xxs +blk.36.attn_k.weight iq3_xxs +blk.36.attn_q.weight iq3_xxs +blk.36.attn_v.weight q5_K +blk.37.ffn_gate.weight iq3_xxs +blk.37.ffn_up.weight iq3_xxs +blk.37.attn_k.weight iq3_xxs +blk.37.attn_q.weight iq3_xxs +blk.37.attn_v.weight q5_K +blk.38.ffn_gate.weight iq3_xxs +blk.38.ffn_up.weight iq3_xxs +blk.38.attn_k.weight iq3_xxs +blk.38.attn_q.weight iq3_xxs +blk.38.attn_v.weight q5_K +blk.39.ffn_gate.weight iq3_xxs +blk.39.ffn_up.weight iq3_xxs +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q5_K +blk.40.ffn_gate.weight iq3_xxs +blk.40.ffn_up.weight iq3_xxs +blk.40.attn_k.weight iq3_xxs +blk.40.attn_q.weight iq3_xxs +blk.40.attn_v.weight q5_K +blk.41.ffn_gate.weight iq3_xxs +blk.41.ffn_up.weight iq3_xxs +blk.41.attn_k.weight iq3_xxs +blk.41.attn_q.weight iq3_xxs +blk.41.attn_v.weight q5_K +blk.42.ffn_gate.weight iq3_xxs +blk.42.ffn_up.weight iq3_xxs +blk.42.attn_k.weight iq3_xxs +blk.42.attn_q.weight iq3_xxs +blk.42.attn_v.weight q5_K +blk.43.ffn_gate.weight iq3_xxs +blk.43.attn_k.weight iq3_xxs +blk.43.attn_q.weight iq3_xxs +blk.43.attn_v.weight q5_K +blk.43.ffn_up.weight iq3_xxs +blk.44.ffn_gate.weight iq3_xxs +blk.44.ffn_up.weight iq3_xxs +blk.44.attn_k.weight iq3_xxs +blk.44.attn_q.weight iq3_xxs +blk.44.attn_v.weight q5_K +blk.45.ffn_gate.weight iq3_xxs +blk.45.ffn_up.weight iq3_xxs +blk.45.attn_k.weight iq3_xxs +blk.45.attn_q.weight iq3_xxs +blk.45.attn_v.weight q5_K +blk.46.attn_k.weight iq3_xxs +blk.46.attn_q.weight iq3_xxs +blk.46.attn_v.weight q5_K +blk.46.ffn_gate.weight iq3_xxs +blk.46.ffn_up.weight iq3_xxs +blk.47.ffn_gate.weight iq3_xxs +blk.47.ffn_up.weight iq3_xxs +blk.47.attn_k.weight iq3_xxs +blk.47.attn_q.weight iq3_xxs +blk.47.attn_v.weight q5_K +blk.48.ffn_gate.weight iq3_xxs +blk.48.ffn_up.weight iq3_xxs +blk.48.attn_k.weight iq3_xxs +blk.48.attn_q.weight iq3_xxs +blk.48.attn_v.weight q5_K +blk.49.attn_k.weight iq3_xxs +blk.49.attn_q.weight iq3_xxs +blk.49.attn_v.weight q5_K +blk.49.ffn_gate.weight iq3_xxs +blk.49.ffn_up.weight iq3_xxs +blk.50.ffn_gate.weight iq3_xxs +blk.50.ffn_up.weight iq3_xxs +blk.50.attn_k.weight iq3_xxs +blk.50.attn_q.weight iq3_xxs +blk.50.attn_v.weight q5_K +blk.51.ffn_gate.weight iq3_xxs +blk.51.ffn_up.weight iq3_xxs +blk.51.attn_k.weight iq3_xxs +blk.51.attn_q.weight iq3_xxs +blk.51.attn_v.weight q5_K +blk.52.ffn_gate.weight iq3_xxs +blk.52.ffn_up.weight iq3_xxs +blk.52.attn_k.weight iq3_xxs +blk.52.attn_q.weight iq3_xxs +blk.52.attn_v.weight q5_K +blk.53.ffn_gate.weight iq3_xxs +blk.53.ffn_up.weight iq3_xxs +blk.53.attn_k.weight iq3_xxs +blk.53.attn_q.weight iq3_xxs +blk.53.attn_v.weight q5_K +blk.54.ffn_gate.weight iq3_xxs +blk.54.ffn_up.weight iq3_xxs +blk.54.attn_k.weight iq3_xxs +blk.54.attn_q.weight iq3_xxs +blk.54.attn_v.weight q5_K +blk.55.ffn_gate.weight iq3_xxs +blk.55.ffn_up.weight iq3_xxs +blk.55.attn_k.weight iq3_xxs +blk.55.attn_q.weight iq3_xxs +blk.55.attn_v.weight q5_K +blk.56.ffn_gate.weight iq3_xxs +blk.56.ffn_up.weight iq3_xxs +blk.56.attn_k.weight iq3_xxs +blk.56.attn_q.weight iq3_xxs +blk.56.attn_v.weight q5_K +blk.57.ffn_gate.weight iq3_xxs +blk.57.attn_k.weight iq3_xxs +blk.57.attn_q.weight iq3_xxs +blk.57.attn_v.weight q5_K +blk.57.ffn_up.weight iq3_xxs +blk.58.ffn_gate.weight iq3_xxs +blk.58.ffn_up.weight iq3_xxs +blk.58.attn_k.weight iq3_xxs +blk.58.attn_q.weight iq3_xxs +blk.58.attn_v.weight q5_K +blk.59.ffn_gate.weight iq3_xxs +blk.59.ffn_up.weight iq3_xxs +blk.59.attn_k.weight iq3_xxs +blk.59.attn_q.weight iq3_xxs +blk.59.attn_v.weight q5_K +blk.60.attn_k.weight iq3_xxs +blk.60.attn_q.weight iq3_xxs +blk.60.attn_v.weight q5_K +blk.60.ffn_gate.weight iq3_xxs +blk.60.ffn_up.weight iq3_xxs +blk.61.ffn_gate.weight iq3_xxs +blk.61.ffn_up.weight iq3_xxs +blk.61.attn_k.weight iq3_xxs +blk.61.attn_q.weight iq3_xxs +blk.61.attn_v.weight q5_K +blk.62.ffn_gate.weight iq3_xxs +blk.62.ffn_up.weight iq3_xxs +blk.62.attn_k.weight iq3_xxs +blk.62.attn_q.weight iq3_xxs +blk.62.attn_v.weight q5_K +blk.63.attn_k.weight iq3_xxs +blk.63.attn_q.weight iq3_xxs +blk.63.attn_v.weight q5_K +blk.63.ffn_gate.weight iq3_xxs +blk.63.ffn_up.weight iq3_xxs +blk.64.ffn_gate.weight iq3_xxs +blk.64.ffn_up.weight iq3_xxs +blk.64.attn_k.weight iq3_xxs +blk.64.attn_q.weight iq3_xxs +blk.64.attn_v.weight q5_K +blk.65.ffn_gate.weight iq3_xxs +blk.65.ffn_up.weight iq3_xxs +blk.65.attn_k.weight iq3_xxs +blk.65.attn_q.weight iq3_xxs +blk.65.attn_v.weight q5_K +blk.66.ffn_gate.weight iq3_xxs +blk.66.ffn_up.weight iq3_xxs +blk.66.attn_k.weight iq3_xxs +blk.66.attn_q.weight iq3_xxs +blk.66.attn_v.weight q5_K +blk.67.ffn_gate.weight iq3_xxs +blk.67.ffn_up.weight iq3_xxs +blk.67.attn_k.weight iq3_xxs +blk.67.attn_q.weight iq3_xxs +blk.67.attn_v.weight q5_K +blk.68.ffn_gate.weight iq3_xxs +blk.68.ffn_up.weight iq3_xxs +blk.68.attn_k.weight iq3_xxs +blk.68.attn_q.weight iq3_xxs +blk.68.attn_v.weight q5_K +blk.69.ffn_gate.weight iq3_xxs +blk.69.ffn_up.weight iq3_xxs +blk.69.attn_k.weight iq3_xxs +blk.69.attn_q.weight iq3_xxs +blk.69.attn_v.weight q5_K +blk.70.attn_k.weight iq3_xxs +blk.70.attn_q.weight iq3_xxs +blk.70.attn_v.weight q5_K +blk.71.attn_k.weight iq3_xxs +blk.71.attn_q.weight iq3_xxs +blk.71.attn_v.weight q5_K +blk.72.attn_k.weight iq3_xxs +blk.72.attn_q.weight iq3_xxs +blk.72.attn_v.weight q5_K +blk.73.attn_k.weight iq3_xxs +blk.73.attn_q.weight iq3_xxs +blk.73.attn_v.weight q5_K +blk.74.attn_k.weight iq3_xxs +blk.74.attn_q.weight iq3_xxs +blk.74.attn_v.weight q5_K +blk.75.attn_k.weight iq3_xxs +blk.75.attn_q.weight iq3_xxs +blk.75.attn_v.weight q5_K +blk.76.attn_k.weight iq3_xxs +blk.76.attn_q.weight iq3_xxs +blk.76.attn_v.weight q5_K +blk.77.attn_k.weight iq3_xxs +blk.77.attn_q.weight iq3_xxs +blk.77.attn_v.weight q5_K +blk.78.attn_k.weight iq3_xxs +blk.78.attn_q.weight iq3_xxs +blk.78.attn_v.weight q5_K +blk.79.attn_k.weight iq3_xxs +blk.79.attn_q.weight iq3_xxs +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ3_XXS] iq3_xxs +token_embd.weight iq3_s +blk.0.ffn_down.weight q4_K +blk.0.attn_k.weight iq2_s +blk.0.attn_output.weight iq3_s +blk.0.attn_q.weight iq2_s +blk.0.attn_v.weight q5_K +blk.1.attn_k.weight iq2_s +blk.1.attn_output.weight iq3_s +blk.1.attn_q.weight iq2_s +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.2.attn_k.weight iq2_s +blk.2.attn_output.weight iq3_s +blk.2.attn_q.weight iq2_s +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q5_K +blk.4.attn_k.weight iq2_s +blk.4.attn_output.weight iq3_s +blk.4.attn_q.weight iq2_s +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.6.attn_k.weight iq2_s +blk.6.attn_output.weight iq3_s +blk.6.attn_q.weight iq2_s +blk.6.attn_v.weight q5_K +blk.7.attn_k.weight iq2_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q4_K +blk.7.attn_output.weight iq3_s +blk.8.ffn_down.weight q4_K +blk.8.attn_k.weight iq2_s +blk.8.attn_output.weight iq3_s +blk.8.attn_q.weight iq2_s +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K +blk.9.attn_k.weight iq2_s +blk.9.attn_output.weight iq3_s +blk.9.attn_q.weight iq2_s +blk.9.attn_v.weight q5_K +blk.10.ffn_down.weight q3_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight q5_K +blk.11.ffn_down.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q5_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight q5_K +blk.14.ffn_down.weight q3_K +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight q5_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K +blk.16.attn_k.weight iq2_s +blk.16.attn_output.weight iq3_s +blk.16.attn_q.weight iq2_s +blk.16.attn_v.weight q5_K +blk.17.ffn_down.weight q3_K +blk.17.attn_k.weight iq2_s +blk.17.attn_output.weight iq3_s +blk.17.attn_q.weight iq2_s +blk.17.attn_v.weight q5_K +blk.18.attn_k.weight iq2_s +blk.18.attn_output.weight iq3_s +blk.18.attn_q.weight iq2_s +blk.18.attn_v.weight q5_K +blk.18.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q5_K +blk.20.ffn_down.weight q3_K +blk.20.attn_k.weight iq2_s +blk.20.attn_output.weight iq3_s +blk.20.attn_q.weight iq2_s +blk.20.attn_v.weight q5_K +blk.21.attn_k.weight iq2_s +blk.21.attn_q.weight iq2_s +blk.21.attn_v.weight q5_K +blk.21.ffn_down.weight q3_K +blk.21.attn_output.weight iq3_s +blk.22.ffn_down.weight q3_K +blk.22.attn_k.weight iq2_s +blk.22.attn_output.weight iq3_s +blk.22.attn_q.weight iq2_s +blk.22.attn_v.weight q5_K +blk.23.ffn_down.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q5_K +blk.24.ffn_down.weight q3_K +blk.24.attn_k.weight iq2_s +blk.24.attn_output.weight iq3_s +blk.24.attn_q.weight iq2_s +blk.24.attn_v.weight q5_K +blk.25.ffn_down.weight q3_K +blk.25.attn_k.weight iq2_s +blk.25.attn_output.weight iq3_s +blk.25.attn_q.weight iq2_s +blk.25.attn_v.weight q5_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight q5_K +blk.26.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q5_K +blk.28.ffn_down.weight q3_K +blk.28.attn_k.weight iq2_s +blk.28.attn_output.weight iq3_s +blk.28.attn_q.weight iq2_s +blk.28.attn_v.weight q5_K +blk.29.attn_k.weight iq2_s +blk.29.attn_output.weight iq3_s +blk.29.attn_q.weight iq2_s +blk.29.attn_v.weight q5_K +blk.29.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K +blk.30.attn_k.weight iq2_s +blk.30.attn_output.weight iq3_s +blk.30.attn_q.weight iq2_s +blk.30.attn_v.weight q5_K +blk.31.ffn_down.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q5_K +blk.32.attn_k.weight iq2_s +blk.32.attn_output.weight iq3_s +blk.32.attn_q.weight iq2_s +blk.32.attn_v.weight q5_K +blk.32.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight q5_K +blk.34.ffn_down.weight q3_K +blk.34.attn_k.weight iq2_s +blk.34.attn_output.weight iq3_s +blk.34.attn_q.weight iq2_s +blk.34.attn_v.weight q5_K +blk.35.attn_k.weight iq2_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q3_K +blk.35.attn_output.weight iq3_s +blk.36.ffn_down.weight q3_K +blk.36.attn_k.weight iq2_s +blk.36.attn_output.weight iq3_s +blk.36.attn_q.weight iq2_s +blk.36.attn_v.weight q5_K +blk.37.ffn_down.weight q3_K +blk.37.attn_k.weight iq2_s +blk.37.attn_output.weight iq3_s +blk.37.attn_q.weight iq2_s +blk.37.attn_v.weight q5_K +blk.38.ffn_down.weight q3_K +blk.38.attn_k.weight iq2_s +blk.38.attn_output.weight iq3_s +blk.38.attn_q.weight iq2_s +blk.38.attn_v.weight q5_K +blk.39.ffn_down.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q5_K +blk.40.attn_k.weight iq2_s +blk.40.attn_output.weight iq3_s +blk.40.attn_q.weight iq2_s +blk.40.attn_v.weight q5_K +blk.40.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K +blk.41.attn_k.weight iq2_s +blk.41.attn_output.weight iq3_s +blk.41.attn_q.weight iq2_s +blk.41.attn_v.weight q5_K +blk.42.ffn_down.weight q3_K +blk.42.attn_k.weight iq2_s +blk.42.attn_output.weight iq3_s +blk.42.attn_q.weight iq2_s +blk.42.attn_v.weight q5_K +blk.43.attn_k.weight iq2_s +blk.43.attn_output.weight iq3_s +blk.43.attn_q.weight iq2_s +blk.43.attn_v.weight q5_K +blk.43.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K +blk.44.attn_k.weight iq2_s +blk.44.attn_output.weight iq3_s +blk.44.attn_q.weight iq2_s +blk.44.attn_v.weight q5_K +blk.45.ffn_down.weight q3_K +blk.45.attn_k.weight iq2_s +blk.45.attn_output.weight iq3_s +blk.45.attn_q.weight iq2_s +blk.45.attn_v.weight q5_K +blk.46.attn_k.weight iq2_s +blk.46.attn_output.weight iq3_s +blk.46.attn_q.weight iq2_s +blk.46.attn_v.weight q5_K +blk.46.ffn_down.weight q3_K +blk.47.ffn_down.weight q3_K +blk.47.attn_k.weight iq2_s +blk.47.attn_output.weight iq3_s +blk.47.attn_q.weight iq2_s +blk.47.attn_v.weight q5_K +blk.48.ffn_down.weight q3_K +blk.48.attn_k.weight iq2_s +blk.48.attn_output.weight iq3_s +blk.48.attn_q.weight iq2_s +blk.48.attn_v.weight q5_K +blk.49.attn_k.weight iq2_s +blk.49.attn_q.weight iq2_s +blk.49.attn_v.weight q5_K +blk.49.ffn_down.weight q3_K +blk.49.attn_output.weight iq3_s +blk.50.ffn_down.weight q3_K +blk.50.attn_k.weight iq2_s +blk.50.attn_output.weight iq3_s +blk.50.attn_q.weight iq2_s +blk.50.attn_v.weight q5_K +blk.51.ffn_down.weight q3_K +blk.51.attn_k.weight iq2_s +blk.51.attn_output.weight iq3_s +blk.51.attn_q.weight iq2_s +blk.51.attn_v.weight q5_K +blk.52.ffn_down.weight q3_K +blk.52.attn_k.weight iq2_s +blk.52.attn_output.weight iq3_s +blk.52.attn_q.weight iq2_s +blk.52.attn_v.weight q5_K +blk.53.ffn_down.weight q3_K +blk.53.attn_k.weight iq2_s +blk.53.attn_output.weight iq3_s +blk.53.attn_q.weight iq2_s +blk.53.attn_v.weight q5_K +blk.54.attn_k.weight iq2_s +blk.54.attn_output.weight iq3_s +blk.54.attn_q.weight iq2_s +blk.54.attn_v.weight q5_K +blk.54.ffn_down.weight q3_K +blk.55.ffn_down.weight q3_K +blk.55.attn_k.weight iq2_s +blk.55.attn_output.weight iq3_s +blk.55.attn_q.weight iq2_s +blk.55.attn_v.weight q5_K +blk.56.ffn_down.weight q3_K +blk.56.attn_k.weight iq2_s +blk.56.attn_output.weight iq3_s +blk.56.attn_q.weight iq2_s +blk.56.attn_v.weight q5_K +blk.57.attn_k.weight iq2_s +blk.57.attn_output.weight iq3_s +blk.57.attn_q.weight iq2_s +blk.57.attn_v.weight q5_K +blk.57.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K +blk.58.attn_k.weight iq2_s +blk.58.attn_output.weight iq3_s +blk.58.attn_q.weight iq2_s +blk.58.attn_v.weight q5_K +blk.59.ffn_down.weight q3_K +blk.59.attn_k.weight iq2_s +blk.59.attn_output.weight iq3_s +blk.59.attn_q.weight iq2_s +blk.59.attn_v.weight q5_K +blk.60.attn_k.weight iq2_s +blk.60.attn_output.weight iq3_s +blk.60.attn_q.weight iq2_s +blk.60.attn_v.weight q5_K +blk.60.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K +blk.61.attn_k.weight iq2_s +blk.61.attn_output.weight iq3_s +blk.61.attn_q.weight iq2_s +blk.61.attn_v.weight q5_K +blk.62.ffn_down.weight q3_K +blk.62.attn_k.weight iq2_s +blk.62.attn_output.weight iq3_s +blk.62.attn_q.weight iq2_s +blk.62.attn_v.weight q5_K +blk.63.attn_k.weight iq2_s +blk.63.attn_q.weight iq2_s +blk.63.attn_v.weight q5_K +blk.63.ffn_down.weight q3_K +blk.63.attn_output.weight iq3_s +blk.64.ffn_down.weight q3_K +blk.64.attn_k.weight iq2_s +blk.64.attn_output.weight iq3_s +blk.64.attn_q.weight iq2_s +blk.64.attn_v.weight q5_K +blk.65.ffn_down.weight q3_K +blk.65.attn_k.weight iq2_s +blk.65.attn_output.weight iq3_s +blk.65.attn_q.weight iq2_s +blk.65.attn_v.weight q5_K +blk.66.ffn_down.weight q3_K +blk.66.attn_k.weight iq2_s +blk.66.attn_output.weight iq3_s +blk.66.attn_q.weight iq2_s +blk.66.attn_v.weight q5_K +blk.67.ffn_down.weight q3_K +blk.67.attn_k.weight iq2_s +blk.67.attn_output.weight iq3_s +blk.67.attn_q.weight iq2_s +blk.67.attn_v.weight q5_K +blk.68.attn_k.weight iq2_s +blk.68.attn_output.weight iq3_s +blk.68.attn_q.weight iq2_s +blk.68.attn_v.weight q5_K +blk.68.ffn_down.weight q3_K +blk.69.ffn_down.weight q3_K +blk.69.attn_k.weight iq2_s +blk.69.attn_output.weight iq3_s +blk.69.attn_q.weight iq2_s +blk.69.attn_v.weight q5_K +blk.70.ffn_down.weight q3_K +blk.70.attn_k.weight iq2_s +blk.70.attn_output.weight iq3_s +blk.70.attn_q.weight iq2_s +blk.70.attn_v.weight q5_K +blk.71.attn_k.weight iq2_s +blk.71.attn_output.weight iq3_s +blk.71.attn_q.weight iq2_s +blk.71.attn_v.weight q5_K +blk.71.ffn_down.weight q3_K +blk.72.ffn_down.weight q3_K +blk.72.attn_k.weight iq2_s +blk.72.attn_output.weight iq3_s +blk.72.attn_q.weight iq2_s +blk.72.attn_v.weight q5_K +blk.73.ffn_down.weight q3_K +blk.73.attn_k.weight iq2_s +blk.73.attn_output.weight iq3_s +blk.73.attn_q.weight iq2_s +blk.73.attn_v.weight q5_K +blk.74.attn_k.weight iq2_s +blk.74.attn_output.weight iq3_s +blk.74.attn_q.weight iq2_s +blk.74.attn_v.weight q5_K +blk.74.ffn_down.weight q3_K +blk.75.ffn_down.weight q3_K +blk.75.attn_k.weight iq2_s +blk.75.attn_output.weight iq3_s +blk.75.attn_q.weight iq2_s +blk.75.attn_v.weight q5_K +blk.76.ffn_down.weight q3_K +blk.76.attn_k.weight iq2_s +blk.76.attn_output.weight iq3_s +blk.76.attn_q.weight iq2_s +blk.76.attn_v.weight q5_K +blk.77.attn_k.weight iq2_s +blk.77.attn_q.weight iq2_s +blk.77.attn_v.weight q5_K +blk.77.ffn_down.weight q3_K +blk.77.attn_output.weight iq3_s +blk.78.ffn_down.weight q3_K +blk.78.attn_k.weight iq2_s +blk.78.attn_output.weight iq3_s +blk.78.attn_q.weight iq2_s +blk.78.attn_v.weight q5_K +blk.79.ffn_down.weight q3_K +blk.79.attn_k.weight iq2_s +blk.79.attn_output.weight iq3_s +blk.79.attn_q.weight iq2_s +blk.79.attn_v.weight q5_K +output.weight q5_K + +[IQ1_S] iq1_s +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.8.ffn_down.weight q2_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq2_xxs +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq2_xxs +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq2_xxs +blk.44.attn_v.weight q4_K +blk.45.attn_output.weight iq2_xxs +blk.45.attn_v.weight q4_K +blk.46.attn_output.weight iq2_xxs +blk.46.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.48.attn_output.weight iq2_xxs +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.49.attn_output.weight iq2_xxs +blk.50.attn_output.weight iq2_xxs +blk.50.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.52.attn_output.weight iq2_xxs +blk.52.attn_v.weight q4_K +blk.53.attn_output.weight iq2_xxs +blk.53.attn_v.weight q4_K +blk.54.attn_output.weight iq2_xxs +blk.54.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.56.attn_output.weight iq2_xxs +blk.56.attn_v.weight q4_K +blk.57.attn_output.weight iq2_xxs +blk.57.attn_v.weight q4_K +blk.58.attn_output.weight iq2_xxs +blk.58.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K +blk.60.attn_output.weight iq2_xxs +blk.60.attn_v.weight q4_K +blk.61.attn_output.weight iq2_xxs +blk.61.attn_v.weight q4_K +blk.62.attn_output.weight iq2_xxs +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.63.attn_output.weight iq2_xxs +blk.64.attn_output.weight iq2_xxs +blk.64.attn_v.weight q4_K +blk.65.attn_output.weight iq2_xxs +blk.65.attn_v.weight q4_K +blk.66.attn_output.weight iq2_xxs +blk.66.attn_v.weight q4_K +blk.67.attn_output.weight iq2_xxs +blk.67.attn_v.weight q4_K +blk.68.attn_output.weight iq2_xxs +blk.68.attn_v.weight q4_K +blk.69.attn_output.weight iq2_xxs +blk.69.attn_v.weight q4_K +blk.70.attn_output.weight iq2_xxs +blk.70.attn_v.weight q4_K +blk.71.attn_output.weight iq2_xxs +blk.71.attn_v.weight q4_K +blk.72.attn_output.weight iq2_xxs +blk.72.attn_v.weight q4_K +blk.73.attn_output.weight iq2_xxs +blk.73.attn_v.weight q4_K +blk.74.attn_output.weight iq2_xxs +blk.74.attn_v.weight q4_K +blk.75.attn_output.weight iq2_xxs +blk.75.attn_v.weight q4_K +blk.76.attn_output.weight iq2_xxs +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.77.attn_output.weight iq2_xxs +blk.78.attn_output.weight iq2_xxs +blk.78.attn_v.weight q4_K +blk.79.attn_output.weight iq2_xxs +blk.79.attn_v.weight q4_K +output.weight q5_K + +[IQ4_NL] iq4_nl +blk.0.ffn_down.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ3_S] iq3_s +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ3_M] iq3_s +blk.0.ffn_down.weight q4_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q5_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.8.ffn_down.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q5_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q5_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q5_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q5_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q5_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q5_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q5_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q5_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q5_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q5_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q5_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.attn_output.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q5_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q5_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q5_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q5_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q5_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q5_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q5_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q5_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q5_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q5_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q5_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q5_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.attn_output.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q5_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q5_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q5_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q5_K +blk.40.attn_output.weight q4_K +blk.40.attn_v.weight q5_K +blk.41.attn_output.weight q4_K +blk.41.attn_v.weight q5_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q5_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q5_K +blk.44.attn_output.weight q4_K +blk.44.attn_v.weight q5_K +blk.45.attn_output.weight q4_K +blk.45.attn_v.weight q5_K +blk.46.attn_output.weight q4_K +blk.46.attn_v.weight q5_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q5_K +blk.48.attn_output.weight q4_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.49.attn_output.weight q4_K +blk.50.attn_output.weight q4_K +blk.50.attn_v.weight q5_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q5_K +blk.52.attn_output.weight q4_K +blk.52.attn_v.weight q5_K +blk.53.attn_output.weight q4_K +blk.53.attn_v.weight q5_K +blk.54.attn_output.weight q4_K +blk.54.attn_v.weight q5_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q5_K +blk.56.attn_output.weight q4_K +blk.56.attn_v.weight q5_K +blk.57.attn_output.weight q4_K +blk.57.attn_v.weight q5_K +blk.58.attn_output.weight q4_K +blk.58.attn_v.weight q5_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q5_K +blk.60.attn_output.weight q4_K +blk.60.attn_v.weight q5_K +blk.61.attn_output.weight q4_K +blk.61.attn_v.weight q5_K +blk.62.attn_output.weight q4_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.63.attn_output.weight q4_K +blk.64.attn_output.weight q4_K +blk.64.attn_v.weight q5_K +blk.65.attn_output.weight q4_K +blk.65.attn_v.weight q5_K +blk.66.attn_output.weight q4_K +blk.66.attn_v.weight q5_K +blk.67.attn_output.weight q4_K +blk.67.attn_v.weight q5_K +blk.68.attn_output.weight q4_K +blk.68.attn_v.weight q5_K +blk.69.attn_output.weight q4_K +blk.69.attn_v.weight q5_K +blk.70.attn_output.weight q4_K +blk.70.attn_v.weight q5_K +blk.71.attn_output.weight q4_K +blk.71.attn_v.weight q5_K +blk.72.attn_output.weight q4_K +blk.72.attn_v.weight q5_K +blk.73.attn_output.weight q4_K +blk.73.attn_v.weight q5_K +blk.74.attn_output.weight q4_K +blk.74.attn_v.weight q5_K +blk.75.attn_output.weight q4_K +blk.75.attn_v.weight q5_K +blk.76.attn_output.weight q4_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.77.attn_output.weight q4_K +blk.78.attn_output.weight q4_K +blk.78.attn_v.weight q5_K +blk.79.attn_output.weight q4_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ2_S] iq2_xs +token_embd.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.8.ffn_down.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq3_s +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq3_s +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq3_s +blk.44.attn_v.weight q4_K +blk.45.attn_output.weight iq3_s +blk.45.attn_v.weight q4_K +blk.46.attn_output.weight iq3_s +blk.46.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.48.attn_output.weight iq3_s +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.49.attn_output.weight iq3_s +blk.50.attn_output.weight iq3_s +blk.50.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.52.attn_output.weight iq3_s +blk.52.attn_v.weight q4_K +blk.53.attn_output.weight iq3_s +blk.53.attn_v.weight q4_K +blk.54.attn_output.weight iq3_s +blk.54.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.56.attn_output.weight iq3_s +blk.56.attn_v.weight q4_K +blk.57.attn_output.weight iq3_s +blk.57.attn_v.weight q4_K +blk.58.attn_output.weight iq3_s +blk.58.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K +blk.60.attn_output.weight iq3_s +blk.60.attn_v.weight q4_K +blk.61.attn_output.weight iq3_s +blk.61.attn_v.weight q4_K +blk.62.attn_output.weight iq3_s +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.63.attn_output.weight iq3_s +blk.64.attn_output.weight iq3_s +blk.64.attn_v.weight q4_K +blk.65.attn_output.weight iq3_s +blk.65.attn_v.weight q4_K +blk.66.attn_output.weight iq3_s +blk.66.attn_v.weight q4_K +blk.67.attn_output.weight iq3_s +blk.67.attn_v.weight q4_K +blk.68.attn_output.weight iq3_s +blk.68.attn_v.weight q4_K +blk.69.attn_output.weight iq3_s +blk.69.attn_v.weight q4_K +blk.70.attn_output.weight iq3_s +blk.70.attn_v.weight q4_K +blk.71.attn_output.weight iq3_s +blk.71.attn_v.weight q4_K +blk.72.attn_output.weight iq3_s +blk.72.attn_v.weight q4_K +blk.73.attn_output.weight iq3_s +blk.73.attn_v.weight q4_K +blk.74.attn_output.weight iq3_s +blk.74.attn_v.weight q4_K +blk.75.attn_output.weight iq3_s +blk.75.attn_v.weight q4_K +blk.76.attn_output.weight iq3_s +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.77.attn_output.weight iq3_s +blk.78.attn_output.weight iq3_s +blk.78.attn_v.weight q4_K +blk.79.attn_output.weight iq3_s +blk.79.attn_v.weight q4_K +output.weight q5_K + +[IQ2_M] iq2_s +token_embd.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight iq3_s +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.8.ffn_down.weight iq3_s +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight iq3_s +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq3_s +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq3_s +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq3_s +blk.44.attn_v.weight q4_K +blk.45.attn_output.weight iq3_s +blk.45.attn_v.weight q4_K +blk.46.attn_output.weight iq3_s +blk.46.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.48.attn_output.weight iq3_s +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.49.attn_output.weight iq3_s +blk.50.attn_output.weight iq3_s +blk.50.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.52.attn_output.weight iq3_s +blk.52.attn_v.weight q4_K +blk.53.attn_output.weight iq3_s +blk.53.attn_v.weight q4_K +blk.54.attn_output.weight iq3_s +blk.54.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.56.attn_output.weight iq3_s +blk.56.attn_v.weight q4_K +blk.57.attn_output.weight iq3_s +blk.57.attn_v.weight q4_K +blk.58.attn_output.weight iq3_s +blk.58.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K +blk.60.attn_output.weight iq3_s +blk.60.attn_v.weight q4_K +blk.61.attn_output.weight iq3_s +blk.61.attn_v.weight q4_K +blk.62.attn_output.weight iq3_s +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.63.attn_output.weight iq3_s +blk.64.attn_output.weight iq3_s +blk.64.attn_v.weight q4_K +blk.65.attn_output.weight iq3_s +blk.65.attn_v.weight q4_K +blk.66.attn_output.weight iq3_s +blk.66.attn_v.weight q4_K +blk.67.attn_output.weight iq3_s +blk.67.attn_v.weight q4_K +blk.68.attn_output.weight iq3_s +blk.68.attn_v.weight q4_K +blk.69.attn_output.weight iq3_s +blk.69.attn_v.weight q4_K +blk.70.attn_output.weight iq3_s +blk.70.attn_v.weight q4_K +blk.71.attn_output.weight iq3_s +blk.71.attn_v.weight q4_K +blk.72.attn_output.weight iq3_s +blk.72.attn_v.weight q4_K +blk.73.attn_output.weight iq3_s +blk.73.attn_v.weight q4_K +blk.74.attn_output.weight iq3_s +blk.74.attn_v.weight q4_K +blk.75.attn_output.weight iq3_s +blk.75.attn_v.weight q4_K +blk.76.attn_output.weight iq3_s +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.77.attn_output.weight iq3_s +blk.78.attn_output.weight iq3_s +blk.78.attn_v.weight q4_K +blk.79.attn_output.weight iq3_s +blk.79.attn_v.weight q4_K +output.weight q5_K + +[IQ4_XS] iq4_xs +blk.0.ffn_down.weight q5_K +blk.0.attn_v.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K +blk.45.attn_v.weight q5_K +blk.46.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.48.attn_v.weight q5_K +blk.49.attn_v.weight q5_K +blk.50.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.52.attn_v.weight q5_K +blk.53.attn_v.weight q5_K +blk.54.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.56.attn_v.weight q5_K +blk.57.attn_v.weight q5_K +blk.58.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.60.attn_v.weight q5_K +blk.61.attn_v.weight q5_K +blk.62.attn_v.weight q5_K +blk.63.attn_v.weight q5_K +blk.64.attn_v.weight q5_K +blk.65.attn_v.weight q5_K +blk.66.attn_v.weight q5_K +blk.67.attn_v.weight q5_K +blk.68.attn_v.weight q5_K +blk.69.attn_v.weight q5_K +blk.70.attn_v.weight q5_K +blk.71.attn_v.weight q5_K +blk.72.attn_v.weight q5_K +blk.73.attn_v.weight q5_K +blk.74.attn_v.weight q5_K +blk.75.attn_v.weight q5_K +blk.76.attn_v.weight q5_K +blk.77.attn_v.weight q5_K +blk.78.attn_v.weight q5_K +blk.79.attn_v.weight q5_K +output.weight q6_K + +[IQ1_M] iq1_m +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.8.ffn_down.weight q2_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq2_xxs +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq2_xxs +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq2_xxs +blk.44.attn_v.weight q4_K +blk.45.attn_output.weight iq2_xxs +blk.45.attn_v.weight q4_K +blk.46.attn_output.weight iq2_xxs +blk.46.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.48.attn_output.weight iq2_xxs +blk.48.attn_v.weight q4_K +blk.49.attn_v.weight q4_K +blk.49.attn_output.weight iq2_xxs +blk.50.attn_output.weight iq2_xxs +blk.50.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.52.attn_output.weight iq2_xxs +blk.52.attn_v.weight q4_K +blk.53.attn_output.weight iq2_xxs +blk.53.attn_v.weight q4_K +blk.54.attn_output.weight iq2_xxs +blk.54.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.56.attn_output.weight iq2_xxs +blk.56.attn_v.weight q4_K +blk.57.attn_output.weight iq2_xxs +blk.57.attn_v.weight q4_K +blk.58.attn_output.weight iq2_xxs +blk.58.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K +blk.60.attn_output.weight iq2_xxs +blk.60.attn_v.weight q4_K +blk.61.attn_output.weight iq2_xxs +blk.61.attn_v.weight q4_K +blk.62.attn_output.weight iq2_xxs +blk.62.attn_v.weight q4_K +blk.63.attn_v.weight q4_K +blk.63.attn_output.weight iq2_xxs +blk.64.attn_output.weight iq2_xxs +blk.64.attn_v.weight q4_K +blk.65.attn_output.weight iq2_xxs +blk.65.attn_v.weight q4_K +blk.66.attn_output.weight iq2_xxs +blk.66.attn_v.weight q4_K +blk.67.attn_output.weight iq2_xxs +blk.67.attn_v.weight q4_K +blk.68.attn_output.weight iq2_xxs +blk.68.attn_v.weight q4_K +blk.69.attn_output.weight iq2_xxs +blk.69.attn_v.weight q4_K +blk.70.attn_output.weight iq2_xxs +blk.70.attn_v.weight q4_K +blk.71.attn_output.weight iq2_xxs +blk.71.attn_v.weight q4_K +blk.72.attn_output.weight iq2_xxs +blk.72.attn_v.weight q4_K +blk.73.attn_output.weight iq2_xxs +blk.73.attn_v.weight q4_K +blk.74.attn_output.weight iq2_xxs +blk.74.attn_v.weight q4_K +blk.75.attn_output.weight iq2_xxs +blk.75.attn_v.weight q4_K +blk.76.attn_output.weight iq2_xxs +blk.76.attn_v.weight q4_K +blk.77.attn_v.weight q4_K +blk.77.attn_output.weight iq2_xxs +blk.78.attn_output.weight iq2_xxs +blk.78.attn_v.weight q4_K +blk.79.attn_output.weight iq2_xxs +blk.79.attn_v.weight q4_K +output.weight q5_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +token_embd.weight q4_K +output.weight q6_K + +[TQ2_0] tq2_0 +token_embd.weight q4_K +output.weight q6_K + +[MXFP4_MOE] mxfp4 +token_embd.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.0.attn_k.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q.weight q8_0 +blk.0.attn_v.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.attn_k.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q.weight q8_0 +blk.1.attn_v.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.2.attn_k.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q.weight q8_0 +blk.2.attn_v.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.4.attn_k.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q.weight q8_0 +blk.4.attn_v.weight q8_0 +blk.4.ffn_down.weight q8_0 +blk.4.ffn_gate.weight q8_0 +blk.4.ffn_up.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 +blk.6.attn_k.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q.weight q8_0 +blk.6.attn_v.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.7.ffn_down.weight q8_0 +blk.7.ffn_gate.weight q8_0 +blk.7.ffn_up.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 +blk.8.attn_k.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q.weight q8_0 +blk.8.attn_v.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 +blk.9.attn_k.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q.weight q8_0 +blk.9.attn_v.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.12.ffn_down.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.14.ffn_down.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 +blk.15.ffn_gate.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.15.ffn_down.weight q8_0 +blk.15.ffn_up.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 +blk.16.attn_k.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q.weight q8_0 +blk.16.attn_v.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 +blk.17.attn_k.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q.weight q8_0 +blk.17.attn_v.weight q8_0 +blk.18.attn_k.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q.weight q8_0 +blk.18.attn_v.weight q8_0 +blk.18.ffn_down.weight q8_0 +blk.18.ffn_gate.weight q8_0 +blk.18.ffn_up.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 +blk.20.attn_k.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q.weight q8_0 +blk.20.attn_v.weight q8_0 +blk.21.attn_k.weight q8_0 +blk.21.attn_q.weight q8_0 +blk.21.attn_v.weight q8_0 +blk.21.ffn_down.weight q8_0 +blk.21.ffn_gate.weight q8_0 +blk.21.ffn_up.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 +blk.22.attn_k.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q.weight q8_0 +blk.22.attn_v.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 +blk.24.attn_k.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q.weight q8_0 +blk.24.attn_v.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 +blk.25.attn_k.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q.weight q8_0 +blk.25.attn_v.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.26.ffn_down.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 +blk.28.attn_k.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q.weight q8_0 +blk.28.attn_v.weight q8_0 +blk.29.ffn_gate.weight q8_0 +blk.29.attn_k.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q.weight q8_0 +blk.29.attn_v.weight q8_0 +blk.29.ffn_down.weight q8_0 +blk.29.ffn_up.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 +blk.30.attn_k.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q.weight q8_0 +blk.30.attn_v.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.32.attn_k.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q.weight q8_0 +blk.32.attn_v.weight q8_0 +blk.32.ffn_down.weight q8_0 +blk.32.ffn_gate.weight q8_0 +blk.32.ffn_up.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 +blk.34.ffn_down.weight q8_0 +blk.34.ffn_gate.weight q8_0 +blk.34.ffn_up.weight q8_0 +blk.34.attn_k.weight q8_0 +blk.34.attn_output.weight q8_0 +blk.34.attn_q.weight q8_0 +blk.34.attn_v.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.35.ffn_down.weight q8_0 +blk.35.ffn_gate.weight q8_0 +blk.35.ffn_up.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.36.ffn_down.weight q8_0 +blk.36.ffn_gate.weight q8_0 +blk.36.ffn_up.weight q8_0 +blk.36.attn_k.weight q8_0 +blk.36.attn_output.weight q8_0 +blk.36.attn_q.weight q8_0 +blk.36.attn_v.weight q8_0 +blk.37.ffn_down.weight q8_0 +blk.37.ffn_gate.weight q8_0 +blk.37.ffn_up.weight q8_0 +blk.37.attn_k.weight q8_0 +blk.37.attn_output.weight q8_0 +blk.37.attn_q.weight q8_0 +blk.37.attn_v.weight q8_0 +blk.38.ffn_down.weight q8_0 +blk.38.ffn_gate.weight q8_0 +blk.38.ffn_up.weight q8_0 +blk.38.attn_k.weight q8_0 +blk.38.attn_output.weight q8_0 +blk.38.attn_q.weight q8_0 +blk.38.attn_v.weight q8_0 +blk.39.ffn_down.weight q8_0 +blk.39.ffn_gate.weight q8_0 +blk.39.ffn_up.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.40.ffn_gate.weight q8_0 +blk.40.ffn_up.weight q8_0 +blk.40.attn_k.weight q8_0 +blk.40.attn_output.weight q8_0 +blk.40.attn_q.weight q8_0 +blk.40.attn_v.weight q8_0 +blk.40.ffn_down.weight q8_0 +blk.41.ffn_down.weight q8_0 +blk.41.ffn_gate.weight q8_0 +blk.41.ffn_up.weight q8_0 +blk.41.attn_k.weight q8_0 +blk.41.attn_output.weight q8_0 +blk.41.attn_q.weight q8_0 +blk.41.attn_v.weight q8_0 +blk.42.ffn_down.weight q8_0 +blk.42.ffn_gate.weight q8_0 +blk.42.ffn_up.weight q8_0 +blk.42.attn_k.weight q8_0 +blk.42.attn_output.weight q8_0 +blk.42.attn_q.weight q8_0 +blk.42.attn_v.weight q8_0 +blk.43.ffn_gate.weight q8_0 +blk.43.attn_k.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q.weight q8_0 +blk.43.attn_v.weight q8_0 +blk.43.ffn_down.weight q8_0 +blk.43.ffn_up.weight q8_0 +blk.44.ffn_down.weight q8_0 +blk.44.ffn_gate.weight q8_0 +blk.44.ffn_up.weight q8_0 +blk.44.attn_k.weight q8_0 +blk.44.attn_output.weight q8_0 +blk.44.attn_q.weight q8_0 +blk.44.attn_v.weight q8_0 +blk.45.ffn_down.weight q8_0 +blk.45.ffn_gate.weight q8_0 +blk.45.ffn_up.weight q8_0 +blk.45.attn_k.weight q8_0 +blk.45.attn_output.weight q8_0 +blk.45.attn_q.weight q8_0 +blk.45.attn_v.weight q8_0 +blk.46.attn_k.weight q8_0 +blk.46.attn_output.weight q8_0 +blk.46.attn_q.weight q8_0 +blk.46.attn_v.weight q8_0 +blk.46.ffn_down.weight q8_0 +blk.46.ffn_gate.weight q8_0 +blk.46.ffn_up.weight q8_0 +blk.47.ffn_down.weight q8_0 +blk.47.ffn_gate.weight q8_0 +blk.47.ffn_up.weight q8_0 +blk.47.attn_k.weight q8_0 +blk.47.attn_output.weight q8_0 +blk.47.attn_q.weight q8_0 +blk.47.attn_v.weight q8_0 +blk.48.ffn_down.weight q8_0 +blk.48.ffn_gate.weight q8_0 +blk.48.ffn_up.weight q8_0 +blk.48.attn_k.weight q8_0 +blk.48.attn_output.weight q8_0 +blk.48.attn_q.weight q8_0 +blk.48.attn_v.weight q8_0 +blk.49.attn_k.weight q8_0 +blk.49.attn_q.weight q8_0 +blk.49.attn_v.weight q8_0 +blk.49.ffn_down.weight q8_0 +blk.49.ffn_gate.weight q8_0 +blk.49.ffn_up.weight q8_0 +blk.49.attn_output.weight q8_0 +blk.50.ffn_down.weight q8_0 +blk.50.ffn_gate.weight q8_0 +blk.50.ffn_up.weight q8_0 +blk.50.attn_k.weight q8_0 +blk.50.attn_output.weight q8_0 +blk.50.attn_q.weight q8_0 +blk.50.attn_v.weight q8_0 +blk.51.ffn_down.weight q8_0 +blk.51.ffn_gate.weight q8_0 +blk.51.ffn_up.weight q8_0 +blk.51.attn_k.weight q8_0 +blk.51.attn_output.weight q8_0 +blk.51.attn_q.weight q8_0 +blk.51.attn_v.weight q8_0 +blk.52.ffn_down.weight q8_0 +blk.52.ffn_gate.weight q8_0 +blk.52.ffn_up.weight q8_0 +blk.52.attn_k.weight q8_0 +blk.52.attn_output.weight q8_0 +blk.52.attn_q.weight q8_0 +blk.52.attn_v.weight q8_0 +blk.53.ffn_down.weight q8_0 +blk.53.ffn_gate.weight q8_0 +blk.53.ffn_up.weight q8_0 +blk.53.attn_k.weight q8_0 +blk.53.attn_output.weight q8_0 +blk.53.attn_q.weight q8_0 +blk.53.attn_v.weight q8_0 +blk.54.ffn_gate.weight q8_0 +blk.54.ffn_up.weight q8_0 +blk.54.attn_k.weight q8_0 +blk.54.attn_output.weight q8_0 +blk.54.attn_q.weight q8_0 +blk.54.attn_v.weight q8_0 +blk.54.ffn_down.weight q8_0 +blk.55.ffn_down.weight q8_0 +blk.55.ffn_gate.weight q8_0 +blk.55.ffn_up.weight q8_0 +blk.55.attn_k.weight q8_0 +blk.55.attn_output.weight q8_0 +blk.55.attn_q.weight q8_0 +blk.55.attn_v.weight q8_0 +blk.56.ffn_down.weight q8_0 +blk.56.ffn_gate.weight q8_0 +blk.56.ffn_up.weight q8_0 +blk.56.attn_k.weight q8_0 +blk.56.attn_output.weight q8_0 +blk.56.attn_q.weight q8_0 +blk.56.attn_v.weight q8_0 +blk.57.ffn_gate.weight q8_0 +blk.57.attn_k.weight q8_0 +blk.57.attn_output.weight q8_0 +blk.57.attn_q.weight q8_0 +blk.57.attn_v.weight q8_0 +blk.57.ffn_down.weight q8_0 +blk.57.ffn_up.weight q8_0 +blk.58.ffn_down.weight q8_0 +blk.58.ffn_gate.weight q8_0 +blk.58.ffn_up.weight q8_0 +blk.58.attn_k.weight q8_0 +blk.58.attn_output.weight q8_0 +blk.58.attn_q.weight q8_0 +blk.58.attn_v.weight q8_0 +blk.59.ffn_down.weight q8_0 +blk.59.ffn_gate.weight q8_0 +blk.59.ffn_up.weight q8_0 +blk.59.attn_k.weight q8_0 +blk.59.attn_output.weight q8_0 +blk.59.attn_q.weight q8_0 +blk.59.attn_v.weight q8_0 +blk.60.attn_k.weight q8_0 +blk.60.attn_output.weight q8_0 +blk.60.attn_q.weight q8_0 +blk.60.attn_v.weight q8_0 +blk.60.ffn_down.weight q8_0 +blk.60.ffn_gate.weight q8_0 +blk.60.ffn_up.weight q8_0 +blk.61.ffn_down.weight q8_0 +blk.61.ffn_gate.weight q8_0 +blk.61.ffn_up.weight q8_0 +blk.61.attn_k.weight q8_0 +blk.61.attn_output.weight q8_0 +blk.61.attn_q.weight q8_0 +blk.61.attn_v.weight q8_0 +blk.62.ffn_down.weight q8_0 +blk.62.ffn_gate.weight q8_0 +blk.62.ffn_up.weight q8_0 +blk.62.attn_k.weight q8_0 +blk.62.attn_output.weight q8_0 +blk.62.attn_q.weight q8_0 +blk.62.attn_v.weight q8_0 +blk.63.attn_k.weight q8_0 +blk.63.attn_q.weight q8_0 +blk.63.attn_v.weight q8_0 +blk.63.ffn_down.weight q8_0 +blk.63.ffn_gate.weight q8_0 +blk.63.ffn_up.weight q8_0 +blk.63.attn_output.weight q8_0 +blk.64.ffn_down.weight q8_0 +blk.64.ffn_gate.weight q8_0 +blk.64.ffn_up.weight q8_0 +blk.64.attn_k.weight q8_0 +blk.64.attn_output.weight q8_0 +blk.64.attn_q.weight q8_0 +blk.64.attn_v.weight q8_0 +blk.65.ffn_down.weight q8_0 +blk.65.ffn_gate.weight q8_0 +blk.65.ffn_up.weight q8_0 +blk.65.attn_k.weight q8_0 +blk.65.attn_output.weight q8_0 +blk.65.attn_q.weight q8_0 +blk.65.attn_v.weight q8_0 +blk.66.ffn_down.weight q8_0 +blk.66.ffn_gate.weight q8_0 +blk.66.ffn_up.weight q8_0 +blk.66.attn_k.weight q8_0 +blk.66.attn_output.weight q8_0 +blk.66.attn_q.weight q8_0 +blk.66.attn_v.weight q8_0 +blk.67.ffn_down.weight q8_0 +blk.67.ffn_gate.weight q8_0 +blk.67.ffn_up.weight q8_0 +blk.67.attn_k.weight q8_0 +blk.67.attn_output.weight q8_0 +blk.67.attn_q.weight q8_0 +blk.67.attn_v.weight q8_0 +blk.68.ffn_gate.weight q8_0 +blk.68.ffn_up.weight q8_0 +blk.68.attn_k.weight q8_0 +blk.68.attn_output.weight q8_0 +blk.68.attn_q.weight q8_0 +blk.68.attn_v.weight q8_0 +blk.68.ffn_down.weight q8_0 +blk.69.ffn_down.weight q8_0 +blk.69.ffn_gate.weight q8_0 +blk.69.ffn_up.weight q8_0 +blk.69.attn_k.weight q8_0 +blk.69.attn_output.weight q8_0 +blk.69.attn_q.weight q8_0 +blk.69.attn_v.weight q8_0 +blk.70.ffn_down.weight q8_0 +blk.70.ffn_gate.weight q8_0 +blk.70.ffn_up.weight q8_0 +blk.70.attn_k.weight q8_0 +blk.70.attn_output.weight q8_0 +blk.70.attn_q.weight q8_0 +blk.70.attn_v.weight q8_0 +blk.71.ffn_gate.weight q8_0 +blk.71.attn_k.weight q8_0 +blk.71.attn_output.weight q8_0 +blk.71.attn_q.weight q8_0 +blk.71.attn_v.weight q8_0 +blk.71.ffn_down.weight q8_0 +blk.71.ffn_up.weight q8_0 +blk.72.ffn_down.weight q8_0 +blk.72.ffn_gate.weight q8_0 +blk.72.ffn_up.weight q8_0 +blk.72.attn_k.weight q8_0 +blk.72.attn_output.weight q8_0 +blk.72.attn_q.weight q8_0 +blk.72.attn_v.weight q8_0 +blk.73.ffn_down.weight q8_0 +blk.73.ffn_gate.weight q8_0 +blk.73.ffn_up.weight q8_0 +blk.73.attn_k.weight q8_0 +blk.73.attn_output.weight q8_0 +blk.73.attn_q.weight q8_0 +blk.73.attn_v.weight q8_0 +blk.74.attn_k.weight q8_0 +blk.74.attn_output.weight q8_0 +blk.74.attn_q.weight q8_0 +blk.74.attn_v.weight q8_0 +blk.74.ffn_down.weight q8_0 +blk.74.ffn_gate.weight q8_0 +blk.74.ffn_up.weight q8_0 +blk.75.ffn_down.weight q8_0 +blk.75.ffn_gate.weight q8_0 +blk.75.ffn_up.weight q8_0 +blk.75.attn_k.weight q8_0 +blk.75.attn_output.weight q8_0 +blk.75.attn_q.weight q8_0 +blk.75.attn_v.weight q8_0 +blk.76.ffn_down.weight q8_0 +blk.76.ffn_gate.weight q8_0 +blk.76.ffn_up.weight q8_0 +blk.76.attn_k.weight q8_0 +blk.76.attn_output.weight q8_0 +blk.76.attn_q.weight q8_0 +blk.76.attn_v.weight q8_0 +blk.77.attn_k.weight q8_0 +blk.77.attn_q.weight q8_0 +blk.77.attn_v.weight q8_0 +blk.77.ffn_down.weight q8_0 +blk.77.ffn_gate.weight q8_0 +blk.77.ffn_up.weight q8_0 +blk.77.attn_output.weight q8_0 +blk.78.ffn_down.weight q8_0 +blk.78.ffn_gate.weight q8_0 +blk.78.ffn_up.weight q8_0 +blk.78.attn_k.weight q8_0 +blk.78.attn_output.weight q8_0 +blk.78.attn_q.weight q8_0 +blk.78.attn_v.weight q8_0 +blk.79.ffn_down.weight q8_0 +blk.79.ffn_gate.weight q8_0 +blk.79.ffn_up.weight q8_0 +blk.79.attn_k.weight q8_0 +blk.79.attn_output.weight q8_0 +blk.79.attn_q.weight q8_0 +blk.79.attn_v.weight q8_0 +output.weight q8_0 diff --git a/tests/snapshots/nemotron-nano-3-30b-a3b.schema b/tests/snapshots/nemotron-nano-3-30b-a3b.schema new file mode 100644 index 0000000000..c68a6e0892 --- /dev/null +++ b/tests/snapshots/nemotron-nano-3-30b-a3b.schema @@ -0,0 +1,696 @@ +# Model: Nemotron-Nano-3-30B-A3B +# n_embd=2688, n_ff=0, n_vocab=131072, n_layer=52, n_head=32, n_head_kv=0, n_expert=128 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q3_K +blk.1.ffn_down_shexp.weight q3_K +blk.3.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q3_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q3_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.42.attn_output.weight q3_K +blk.42.attn_v.weight q3_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K + +[Q3_K_S] q3_K +output.weight q8_0 + +[Q3_K_M] q3_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.8.ffn_down_shexp.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.10.ffn_down_shexp.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q5_K +blk.13.ffn_down_exps.weight q4_K +blk.13.ffn_down_shexp.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.17.ffn_down_shexp.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.20.ffn_down_shexp.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.22.ffn_down_shexp.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.24.ffn_down_shexp.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.29.ffn_down_shexp.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.34.ffn_down_shexp.weight q4_K +blk.36.ffn_down_exps.weight q4_K +blk.36.ffn_down_shexp.weight q4_K +blk.38.ffn_down_exps.weight q4_K +blk.38.ffn_down_shexp.weight q4_K +blk.40.ffn_down_exps.weight q4_K +blk.40.ffn_down_shexp.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K +blk.45.ffn_down_exps.weight q4_K +blk.45.ffn_down_shexp.weight q4_K +blk.47.ffn_down_exps.weight q4_K +blk.47.ffn_down_shexp.weight q4_K +blk.49.ffn_down_exps.weight q4_K +blk.49.ffn_down_shexp.weight q4_K +blk.51.ffn_down_exps.weight q4_K +blk.51.ffn_down_shexp.weight q4_K + +[Q3_K_L] q3_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.8.ffn_down_exps.weight q5_K +blk.8.ffn_down_shexp.weight q5_K +blk.10.ffn_down_exps.weight q5_K +blk.10.ffn_down_shexp.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.ffn_down_exps.weight q5_K +blk.13.ffn_down_shexp.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K +blk.17.ffn_down_exps.weight q5_K +blk.17.ffn_down_shexp.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.ffn_down_exps.weight q5_K +blk.20.ffn_down_shexp.weight q5_K +blk.22.ffn_down_exps.weight q5_K +blk.22.ffn_down_shexp.weight q5_K +blk.24.ffn_down_exps.weight q5_K +blk.24.ffn_down_shexp.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K +blk.29.ffn_down_exps.weight q5_K +blk.29.ffn_down_shexp.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.ffn_down_exps.weight q5_K +blk.34.ffn_down_shexp.weight q5_K +blk.36.ffn_down_exps.weight q5_K +blk.36.ffn_down_shexp.weight q5_K +blk.38.ffn_down_exps.weight q5_K +blk.38.ffn_down_shexp.weight q5_K +blk.40.ffn_down_exps.weight q5_K +blk.40.ffn_down_shexp.weight q5_K +blk.42.attn_output.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K +blk.45.ffn_down_exps.weight q5_K +blk.45.ffn_down_shexp.weight q5_K +blk.47.ffn_down_exps.weight q5_K +blk.47.ffn_down_shexp.weight q5_K +blk.49.ffn_down_exps.weight q5_K +blk.49.ffn_down_shexp.weight q5_K +blk.51.ffn_down_exps.weight q5_K +blk.51.ffn_down_shexp.weight q5_K + +[Q4_K_S] q4_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.5.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.26.attn_v.weight q5_K + +[Q4_K_M] q4_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.8.ffn_down_exps.weight q6_K +blk.8.ffn_down_shexp.weight q6_K +blk.17.ffn_down_exps.weight q6_K +blk.17.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.20.ffn_down_exps.weight q6_K +blk.20.ffn_down_shexp.weight q6_K +blk.29.ffn_down_exps.weight q6_K +blk.29.ffn_down_shexp.weight q6_K +blk.38.ffn_down_exps.weight q6_K +blk.38.ffn_down_shexp.weight q6_K +blk.42.attn_v.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.47.ffn_down_exps.weight q6_K +blk.47.ffn_down_shexp.weight q6_K +blk.49.ffn_down_exps.weight q6_K +blk.49.ffn_down_shexp.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K + +[Q5_K_S] q5_K +output.weight q8_0 + +[Q5_K_M] q5_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.8.ffn_down_exps.weight q6_K +blk.8.ffn_down_shexp.weight q6_K +blk.17.ffn_down_exps.weight q6_K +blk.17.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.20.ffn_down_exps.weight q6_K +blk.20.ffn_down_shexp.weight q6_K +blk.29.ffn_down_exps.weight q6_K +blk.29.ffn_down_shexp.weight q6_K +blk.38.ffn_down_exps.weight q6_K +blk.38.ffn_down_shexp.weight q6_K +blk.42.attn_v.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.47.ffn_down_exps.weight q6_K +blk.47.ffn_down_shexp.weight q6_K +blk.49.ffn_down_exps.weight q6_K +blk.49.ffn_down_shexp.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K + +[Q6_K] q6_K +output.weight q8_0 + +[IQ2_XXS] iq2_xxs +output.weight q8_0 +token_embd.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q2_K +blk.6.ffn_down_shexp.weight q2_K +blk.12.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.42.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q8_0 +token_embd.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q2_K +blk.6.ffn_down_shexp.weight q2_K +blk.12.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.42.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q8_0 +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K + +[IQ3_XS] iq3_s +output.weight q8_0 +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs +blk.6.ffn_up_shexp.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs +blk.40.ffn_up_exps.weight iq3_xxs +blk.40.ffn_up_shexp.weight iq3_xxs +blk.42.attn_k.weight iq3_xxs +blk.42.attn_q.weight iq3_xxs +blk.43.ffn_up_exps.weight iq3_xxs +blk.43.ffn_up_shexp.weight iq3_xxs + +[IQ3_XXS] iq3_xxs +output.weight q8_0 +token_embd.weight iq3_s +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight iq3_s +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight iq3_s +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight iq3_s +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight iq3_s +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight iq3_s +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.42.attn_k.weight iq2_s +blk.42.attn_output.weight iq3_s +blk.42.attn_q.weight iq2_s +blk.42.attn_v.weight iq3_s +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K + +[IQ1_S] iq1_s +output.weight q8_0 +token_embd.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q2_K +blk.6.ffn_down_shexp.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K + +[IQ3_S] iq3_s +output.weight q8_0 + +[IQ3_M] iq3_s +output.weight q8_0 +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q8_0 +token_embd.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight iq3_s +blk.6.ffn_down_shexp.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q8_0 +token_embd.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight iq3_s +blk.6.ffn_down_shexp.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q8_0 +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K + +[IQ1_M] iq1_m +output.weight q8_0 +token_embd.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q2_K +blk.6.ffn_down_shexp.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q8_0 +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q8_0 +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.ssm_in.weight q8_0 +blk.0.ssm_out.weight q8_0 +blk.1.ffn_down_shexp.weight q8_0 +blk.1.ffn_up_shexp.weight q8_0 +blk.2.ssm_in.weight q8_0 +blk.2.ssm_out.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 +blk.4.ssm_in.weight q8_0 +blk.4.ssm_out.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 +blk.7.ssm_in.weight q8_0 +blk.7.ssm_out.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 +blk.9.ssm_in.weight q8_0 +blk.9.ssm_out.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 +blk.11.ssm_in.weight q8_0 +blk.11.ssm_out.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 +blk.14.ssm_in.weight q8_0 +blk.14.ssm_out.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 +blk.16.ssm_in.weight q8_0 +blk.16.ssm_out.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 +blk.18.ssm_in.weight q8_0 +blk.18.ssm_out.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 +blk.21.ssm_in.weight q8_0 +blk.21.ssm_out.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 +blk.23.ssm_in.weight q8_0 +blk.23.ssm_out.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 +blk.25.ssm_in.weight q8_0 +blk.25.ssm_out.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 +blk.28.ssm_in.weight q8_0 +blk.28.ssm_out.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 +blk.30.ssm_in.weight q8_0 +blk.30.ssm_out.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 +blk.32.ssm_in.weight q8_0 +blk.32.ssm_out.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 +blk.35.ssm_in.weight q8_0 +blk.35.ssm_out.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 +blk.37.ssm_in.weight q8_0 +blk.37.ssm_out.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 +blk.39.ssm_in.weight q8_0 +blk.39.ssm_out.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 +blk.41.ssm_in.weight q8_0 +blk.41.ssm_out.weight q8_0 +blk.42.attn_k.weight q8_0 +blk.42.attn_output.weight q8_0 +blk.42.attn_q.weight q8_0 +blk.42.attn_v.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 +blk.44.ssm_in.weight q8_0 +blk.44.ssm_out.weight q8_0 +blk.45.ffn_down_shexp.weight q8_0 +blk.45.ffn_up_shexp.weight q8_0 +blk.46.ssm_in.weight q8_0 +blk.46.ssm_out.weight q8_0 +blk.47.ffn_down_shexp.weight q8_0 +blk.47.ffn_up_shexp.weight q8_0 +blk.48.ssm_in.weight q8_0 +blk.48.ssm_out.weight q8_0 +blk.49.ffn_down_shexp.weight q8_0 +blk.49.ffn_up_shexp.weight q8_0 +blk.50.ssm_in.weight q8_0 +blk.50.ssm_out.weight q8_0 +blk.51.ffn_down_shexp.weight q8_0 +blk.51.ffn_up_shexp.weight q8_0 diff --git a/tests/snapshots/qwen3-14b.schema b/tests/snapshots/qwen3-14b.schema new file mode 100644 index 0000000000..010849a57b --- /dev/null +++ b/tests/snapshots/qwen3-14b.schema @@ -0,0 +1,1908 @@ +# Model: Qwen3-14B +# n_embd=5120, n_ff=17408, n_vocab=151936, n_layer=40, n_head=40, n_head_kv=8 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q6_K +blk.0.attn_output.weight q3_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q3_K +blk.1.attn_output.weight q3_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q3_K +blk.4.attn_output.weight q3_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down.weight q3_K +blk.18.attn_output.weight q3_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down.weight q3_K +blk.21.attn_output.weight q3_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down.weight q3_K +blk.29.attn_output.weight q3_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q3_K +blk.32.attn_output.weight q3_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down.weight q3_K +blk.34.attn_output.weight q3_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q3_K +blk.36.attn_output.weight q3_K +blk.36.attn_v.weight q4_K +blk.36.ffn_down.weight q3_K +blk.37.attn_output.weight q3_K +blk.37.attn_v.weight q4_K +blk.37.ffn_down.weight q3_K +blk.38.attn_output.weight q3_K +blk.38.attn_v.weight q4_K +blk.38.ffn_down.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q4_K +blk.36.ffn_down.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q4_K +blk.37.ffn_down.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q4_K +blk.38.ffn_down.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q4_K + +[Q3_K_L] q3_K +output.weight q6_K +blk.0.attn_output.weight q5_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.5.ffn_down.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.8.ffn_down.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.10.ffn_down.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.13.ffn_down.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.14.ffn_down.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.16.ffn_down.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.17.ffn_down.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.20.ffn_down.weight q5_K +blk.21.attn_output.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.22.ffn_down.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.24.ffn_down.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.25.ffn_down.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.28.ffn_down.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.30.ffn_down.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.33.ffn_down.weight q5_K +blk.34.attn_output.weight q5_K +blk.34.attn_v.weight q5_K +blk.34.ffn_down.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q5_K +blk.36.attn_output.weight q5_K +blk.36.attn_v.weight q5_K +blk.36.ffn_down.weight q5_K +blk.37.attn_output.weight q5_K +blk.37.attn_v.weight q5_K +blk.37.ffn_down.weight q5_K +blk.38.attn_output.weight q5_K +blk.38.attn_v.weight q5_K +blk.38.ffn_down.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q5_K + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.ffn_down.weight q5_K + +[Q4_K_M] q4_K +output.weight q6_K +blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.10.attn_v.weight q6_K +blk.10.ffn_down.weight q6_K +blk.13.attn_v.weight q6_K +blk.13.ffn_down.weight q6_K +blk.16.attn_v.weight q6_K +blk.16.ffn_down.weight q6_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down.weight q6_K +blk.22.attn_v.weight q6_K +blk.22.ffn_down.weight q6_K +blk.25.attn_v.weight q6_K +blk.25.ffn_down.weight q6_K +blk.28.attn_v.weight q6_K +blk.28.ffn_down.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down.weight q6_K +blk.36.attn_v.weight q6_K +blk.36.ffn_down.weight q6_K +blk.37.attn_v.weight q6_K +blk.37.ffn_down.weight q6_K +blk.38.attn_v.weight q6_K +blk.38.ffn_down.weight q6_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +output.weight q6_K +blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.10.attn_v.weight q6_K +blk.10.ffn_down.weight q6_K +blk.13.attn_v.weight q6_K +blk.13.ffn_down.weight q6_K +blk.16.attn_v.weight q6_K +blk.16.ffn_down.weight q6_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down.weight q6_K +blk.22.attn_v.weight q6_K +blk.22.ffn_down.weight q6_K +blk.25.attn_v.weight q6_K +blk.25.ffn_down.weight q6_K +blk.28.attn_v.weight q6_K +blk.28.ffn_down.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down.weight q6_K +blk.36.attn_v.weight q6_K +blk.36.ffn_down.weight q6_K +blk.37.attn_v.weight q6_K +blk.37.ffn_down.weight q6_K +blk.38.attn_v.weight q6_K +blk.38.ffn_down.weight q6_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q6_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K + +[IQ3_XS] iq3_s +output.weight q6_K +blk.0.attn_k.weight iq3_xxs +blk.0.attn_q.weight iq3_xxs +blk.0.attn_v.weight q4_K +blk.1.attn_k.weight iq3_xxs +blk.1.attn_q.weight iq3_xxs +blk.1.attn_v.weight q4_K +blk.2.attn_k.weight iq3_xxs +blk.2.attn_q.weight iq3_xxs +blk.2.attn_v.weight q4_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.4.attn_k.weight iq3_xxs +blk.4.attn_q.weight iq3_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.5.attn_v.weight q4_K +blk.5.ffn_gate.weight iq3_xxs +blk.5.ffn_up.weight iq3_xxs +blk.6.attn_k.weight iq3_xxs +blk.6.attn_q.weight iq3_xxs +blk.6.attn_v.weight q4_K +blk.6.ffn_gate.weight iq3_xxs +blk.6.ffn_up.weight iq3_xxs +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_gate.weight iq3_xxs +blk.7.ffn_up.weight iq3_xxs +blk.8.attn_k.weight iq3_xxs +blk.8.attn_q.weight iq3_xxs +blk.8.attn_v.weight q4_K +blk.8.ffn_gate.weight iq3_xxs +blk.8.ffn_up.weight iq3_xxs +blk.9.attn_k.weight iq3_xxs +blk.9.attn_q.weight iq3_xxs +blk.9.attn_v.weight q4_K +blk.9.ffn_gate.weight iq3_xxs +blk.9.ffn_up.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.10.attn_v.weight q4_K +blk.10.ffn_gate.weight iq3_xxs +blk.10.ffn_up.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate.weight iq3_xxs +blk.11.ffn_up.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.12.attn_v.weight q4_K +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.13.attn_v.weight q4_K +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.14.attn_v.weight q4_K +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate.weight iq3_xxs +blk.15.ffn_up.weight iq3_xxs +blk.16.attn_k.weight iq3_xxs +blk.16.attn_q.weight iq3_xxs +blk.16.attn_v.weight q4_K +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs +blk.17.attn_k.weight iq3_xxs +blk.17.attn_q.weight iq3_xxs +blk.17.attn_v.weight q4_K +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs +blk.18.attn_k.weight iq3_xxs +blk.18.attn_q.weight iq3_xxs +blk.18.attn_v.weight q4_K +blk.18.ffn_gate.weight iq3_xxs +blk.18.ffn_up.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs +blk.20.attn_k.weight iq3_xxs +blk.20.attn_q.weight iq3_xxs +blk.20.attn_v.weight q4_K +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs +blk.21.attn_k.weight iq3_xxs +blk.21.attn_q.weight iq3_xxs +blk.21.attn_v.weight q4_K +blk.21.ffn_gate.weight iq3_xxs +blk.21.ffn_up.weight iq3_xxs +blk.22.attn_k.weight iq3_xxs +blk.22.attn_q.weight iq3_xxs +blk.22.attn_v.weight q4_K +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs +blk.24.attn_k.weight iq3_xxs +blk.24.attn_q.weight iq3_xxs +blk.24.attn_v.weight q4_K +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs +blk.25.attn_k.weight iq3_xxs +blk.25.attn_q.weight iq3_xxs +blk.25.attn_v.weight q4_K +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.26.attn_v.weight q4_K +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs +blk.28.attn_k.weight iq3_xxs +blk.28.attn_q.weight iq3_xxs +blk.28.attn_v.weight q4_K +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs +blk.29.attn_k.weight iq3_xxs +blk.29.attn_q.weight iq3_xxs +blk.29.attn_v.weight q4_K +blk.29.ffn_gate.weight iq3_xxs +blk.29.ffn_up.weight iq3_xxs +blk.30.attn_k.weight iq3_xxs +blk.30.attn_q.weight iq3_xxs +blk.30.attn_v.weight q4_K +blk.30.ffn_gate.weight iq3_xxs +blk.30.ffn_up.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_gate.weight iq3_xxs +blk.31.ffn_up.weight iq3_xxs +blk.32.attn_k.weight iq3_xxs +blk.32.attn_q.weight iq3_xxs +blk.32.attn_v.weight q4_K +blk.32.ffn_gate.weight iq3_xxs +blk.32.ffn_up.weight iq3_xxs +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs +blk.33.attn_v.weight q4_K +blk.33.ffn_gate.weight iq3_xxs +blk.33.ffn_up.weight iq3_xxs +blk.34.attn_k.weight iq3_xxs +blk.34.attn_q.weight iq3_xxs +blk.34.attn_v.weight q4_K +blk.34.ffn_gate.weight iq3_xxs +blk.34.ffn_up.weight iq3_xxs +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K +blk.36.attn_k.weight iq3_xxs +blk.36.attn_q.weight iq3_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_k.weight iq3_xxs +blk.37.attn_q.weight iq3_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_k.weight iq3_xxs +blk.38.attn_q.weight iq3_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q4_K + +[IQ3_XXS] iq3_xxs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_k.weight iq2_s +blk.0.attn_output.weight iq3_s +blk.0.attn_q.weight iq2_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_k.weight iq2_s +blk.1.attn_output.weight iq3_s +blk.1.attn_q.weight iq2_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_k.weight iq2_s +blk.2.attn_output.weight iq3_s +blk.2.attn_q.weight iq2_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.attn_k.weight iq2_s +blk.4.attn_output.weight iq3_s +blk.4.attn_q.weight iq2_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight q4_K +blk.5.ffn_down.weight q3_K +blk.6.attn_k.weight iq2_s +blk.6.attn_output.weight iq3_s +blk.6.attn_q.weight iq2_s +blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q3_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q3_K +blk.8.attn_k.weight iq2_s +blk.8.attn_output.weight iq3_s +blk.8.attn_q.weight iq2_s +blk.8.attn_v.weight q4_K +blk.8.ffn_down.weight q3_K +blk.9.attn_k.weight iq2_s +blk.9.attn_output.weight iq3_s +blk.9.attn_q.weight iq2_s +blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q3_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight q4_K +blk.10.ffn_down.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight q4_K +blk.12.ffn_down.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight q4_K +blk.13.ffn_down.weight q3_K +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight q4_K +blk.14.ffn_down.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q3_K +blk.16.attn_k.weight iq2_s +blk.16.attn_output.weight iq3_s +blk.16.attn_q.weight iq2_s +blk.16.attn_v.weight q4_K +blk.16.ffn_down.weight q3_K +blk.17.attn_k.weight iq2_s +blk.17.attn_output.weight iq3_s +blk.17.attn_q.weight iq2_s +blk.17.attn_v.weight q4_K +blk.17.ffn_down.weight q3_K +blk.18.attn_k.weight iq2_s +blk.18.attn_output.weight iq3_s +blk.18.attn_q.weight iq2_s +blk.18.attn_v.weight q4_K +blk.18.ffn_down.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q3_K +blk.20.attn_k.weight iq2_s +blk.20.attn_output.weight iq3_s +blk.20.attn_q.weight iq2_s +blk.20.attn_v.weight q4_K +blk.20.ffn_down.weight q3_K +blk.21.attn_k.weight iq2_s +blk.21.attn_output.weight iq3_s +blk.21.attn_q.weight iq2_s +blk.21.attn_v.weight q4_K +blk.21.ffn_down.weight q3_K +blk.22.attn_k.weight iq2_s +blk.22.attn_output.weight iq3_s +blk.22.attn_q.weight iq2_s +blk.22.attn_v.weight q4_K +blk.22.ffn_down.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q3_K +blk.24.attn_k.weight iq2_s +blk.24.attn_output.weight iq3_s +blk.24.attn_q.weight iq2_s +blk.24.attn_v.weight q4_K +blk.24.ffn_down.weight q3_K +blk.25.attn_k.weight iq2_s +blk.25.attn_output.weight iq3_s +blk.25.attn_q.weight iq2_s +blk.25.attn_v.weight q4_K +blk.25.ffn_down.weight q3_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight q4_K +blk.26.ffn_down.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q3_K +blk.28.attn_k.weight iq2_s +blk.28.attn_output.weight iq3_s +blk.28.attn_q.weight iq2_s +blk.28.attn_v.weight q4_K +blk.28.ffn_down.weight q3_K +blk.29.attn_k.weight iq2_s +blk.29.attn_output.weight iq3_s +blk.29.attn_q.weight iq2_s +blk.29.attn_v.weight q4_K +blk.29.ffn_down.weight q3_K +blk.30.attn_k.weight iq2_s +blk.30.attn_output.weight iq3_s +blk.30.attn_q.weight iq2_s +blk.30.attn_v.weight q4_K +blk.30.ffn_down.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q3_K +blk.32.attn_k.weight iq2_s +blk.32.attn_output.weight iq3_s +blk.32.attn_q.weight iq2_s +blk.32.attn_v.weight q4_K +blk.32.ffn_down.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight q4_K +blk.33.ffn_down.weight q3_K +blk.34.attn_k.weight iq2_s +blk.34.attn_output.weight iq3_s +blk.34.attn_q.weight iq2_s +blk.34.attn_v.weight q4_K +blk.34.ffn_down.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q3_K +blk.36.attn_k.weight iq2_s +blk.36.attn_output.weight iq3_s +blk.36.attn_q.weight iq2_s +blk.36.attn_v.weight q4_K +blk.36.ffn_down.weight q3_K +blk.37.attn_k.weight iq2_s +blk.37.attn_output.weight iq3_s +blk.37.attn_q.weight iq2_s +blk.37.attn_v.weight q4_K +blk.37.ffn_down.weight q3_K +blk.38.attn_k.weight iq2_s +blk.38.attn_output.weight iq3_s +blk.38.attn_q.weight iq2_s +blk.38.attn_v.weight q4_K +blk.38.ffn_down.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q3_K + +[IQ1_S] iq1_s +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K + +[IQ3_S] iq3_s +output.weight q6_K +blk.0.attn_v.weight q4_K +blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K + +[IQ3_M] iq3_s +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight iq3_s +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_down.weight q2_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q6_K +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q6_K +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_k.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q.weight q8_0 +blk.0.attn_v.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.1.attn_k.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q.weight q8_0 +blk.1.attn_v.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.2.attn_k.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q.weight q8_0 +blk.2.attn_v.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 +blk.4.attn_k.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q.weight q8_0 +blk.4.attn_v.weight q8_0 +blk.4.ffn_down.weight q8_0 +blk.4.ffn_gate.weight q8_0 +blk.4.ffn_up.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 +blk.6.attn_k.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q.weight q8_0 +blk.6.attn_v.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.7.ffn_down.weight q8_0 +blk.7.ffn_gate.weight q8_0 +blk.7.ffn_up.weight q8_0 +blk.8.attn_k.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q.weight q8_0 +blk.8.attn_v.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 +blk.9.attn_k.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q.weight q8_0 +blk.9.attn_v.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.12.ffn_down.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 +blk.14.ffn_down.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.15.ffn_down.weight q8_0 +blk.15.ffn_gate.weight q8_0 +blk.15.ffn_up.weight q8_0 +blk.16.attn_k.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q.weight q8_0 +blk.16.attn_v.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 +blk.17.attn_k.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q.weight q8_0 +blk.17.attn_v.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 +blk.18.attn_k.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q.weight q8_0 +blk.18.attn_v.weight q8_0 +blk.18.ffn_down.weight q8_0 +blk.18.ffn_gate.weight q8_0 +blk.18.ffn_up.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 +blk.20.attn_k.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q.weight q8_0 +blk.20.attn_v.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 +blk.21.attn_k.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.21.attn_q.weight q8_0 +blk.21.attn_v.weight q8_0 +blk.21.ffn_down.weight q8_0 +blk.21.ffn_gate.weight q8_0 +blk.21.ffn_up.weight q8_0 +blk.22.attn_k.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q.weight q8_0 +blk.22.attn_v.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 +blk.24.attn_k.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q.weight q8_0 +blk.24.attn_v.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 +blk.25.attn_k.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q.weight q8_0 +blk.25.attn_v.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.26.ffn_down.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 +blk.28.attn_k.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q.weight q8_0 +blk.28.attn_v.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 +blk.29.attn_k.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q.weight q8_0 +blk.29.attn_v.weight q8_0 +blk.29.ffn_down.weight q8_0 +blk.29.ffn_gate.weight q8_0 +blk.29.ffn_up.weight q8_0 +blk.30.attn_k.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q.weight q8_0 +blk.30.attn_v.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 +blk.32.attn_k.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q.weight q8_0 +blk.32.attn_v.weight q8_0 +blk.32.ffn_down.weight q8_0 +blk.32.ffn_gate.weight q8_0 +blk.32.ffn_up.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 +blk.34.attn_k.weight q8_0 +blk.34.attn_output.weight q8_0 +blk.34.attn_q.weight q8_0 +blk.34.attn_v.weight q8_0 +blk.34.ffn_down.weight q8_0 +blk.34.ffn_gate.weight q8_0 +blk.34.ffn_up.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.35.ffn_down.weight q8_0 +blk.35.ffn_gate.weight q8_0 +blk.35.ffn_up.weight q8_0 +blk.36.attn_k.weight q8_0 +blk.36.attn_output.weight q8_0 +blk.36.attn_q.weight q8_0 +blk.36.attn_v.weight q8_0 +blk.36.ffn_down.weight q8_0 +blk.36.ffn_gate.weight q8_0 +blk.36.ffn_up.weight q8_0 +blk.37.attn_k.weight q8_0 +blk.37.attn_output.weight q8_0 +blk.37.attn_q.weight q8_0 +blk.37.attn_v.weight q8_0 +blk.37.ffn_down.weight q8_0 +blk.37.ffn_gate.weight q8_0 +blk.37.ffn_up.weight q8_0 +blk.38.attn_k.weight q8_0 +blk.38.attn_output.weight q8_0 +blk.38.attn_q.weight q8_0 +blk.38.attn_v.weight q8_0 +blk.38.ffn_down.weight q8_0 +blk.38.ffn_gate.weight q8_0 +blk.38.ffn_up.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.39.ffn_down.weight q8_0 +blk.39.ffn_gate.weight q8_0 +blk.39.ffn_up.weight q8_0 diff --git a/tests/snapshots/qwen3-coder-next.schema b/tests/snapshots/qwen3-coder-next.schema new file mode 100644 index 0000000000..d71fe86a1f --- /dev/null +++ b/tests/snapshots/qwen3-coder-next.schema @@ -0,0 +1,1713 @@ +# Model: Qwen3-Coder-Next +# n_embd=2048, n_ff=5120, n_vocab=151936, n_layer=48, n_head=16, n_head_kv=2, n_expert=512 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +blk.0.ffn_down_exps.weight q3_K +blk.0.ffn_down_shexp.weight q3_K +blk.1.ffn_down_shexp.weight q3_K +blk.1.ffn_down_exps.weight q3_K +blk.2.ffn_down_shexp.weight q3_K +blk.2.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q3_K +blk.4.ffn_down_shexp.weight q3_K +blk.4.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K +blk.5.ffn_down_exps.weight q3_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.9.ffn_down_exps.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.14.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.16.ffn_down_exps.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.18.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.21.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.24.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.25.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.26.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.28.ffn_down_exps.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.30.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.32.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.33.ffn_down_exps.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.37.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.41.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.42.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.43.attn_output.weight q3_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.44.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.47.attn_output.weight q3_K +blk.47.attn_v.weight q4_K +output.weight q6_K +blk.47.ffn_down_exps.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down_shexp.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_shexp.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down_shexp.weight q4_K +blk.9.ffn_down_exps.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.11.ffn_down_shexp.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down_exps.weight q4_K +blk.12.ffn_down_shexp.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down_shexp.weight q4_K +blk.13.ffn_down_exps.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down_shexp.weight q4_K +blk.14.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down_shexp.weight q4_K +blk.16.ffn_down_exps.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.17.ffn_down_shexp.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down_shexp.weight q4_K +blk.18.ffn_down_exps.weight q4_K +blk.19.ffn_down_shexp.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down_shexp.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down_shexp.weight q4_K +blk.21.ffn_down_exps.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down_shexp.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.23.ffn_down_shexp.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down_shexp.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down_shexp.weight q4_K +blk.25.ffn_down_exps.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down_shexp.weight q4_K +blk.26.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down_shexp.weight q4_K +blk.28.ffn_down_exps.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.29.ffn_down_shexp.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down_shexp.weight q4_K +blk.30.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down_shexp.weight q4_K +blk.32.ffn_down_exps.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down_shexp.weight q4_K +blk.33.ffn_down_exps.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.34.ffn_down_shexp.weight q4_K +blk.35.ffn_down_shexp.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down_shexp.weight q4_K +blk.36.ffn_down_exps.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down_shexp.weight q4_K +blk.37.ffn_down_exps.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down_shexp.weight q4_K +blk.38.ffn_down_exps.weight q4_K +blk.39.ffn_down_shexp.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down_exps.weight q4_K +blk.40.ffn_down_shexp.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down_shexp.weight q4_K +blk.41.ffn_down_exps.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down_shexp.weight q4_K +blk.42.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down_shexp.weight q4_K +blk.44.ffn_down_exps.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down_shexp.weight q4_K +blk.45.ffn_down_exps.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down_exps.weight q4_K +blk.46.ffn_down_shexp.weight q4_K +blk.47.ffn_down_shexp.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q6_K +blk.47.ffn_down_exps.weight q4_K + +[Q3_K_L] q3_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.ffn_down_shexp.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q5_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down_shexp.weight q5_K +blk.8.ffn_down_exps.weight q5_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_shexp.weight q5_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down_shexp.weight q5_K +blk.9.ffn_down_exps.weight q5_K +blk.10.ffn_down_exps.weight q5_K +blk.11.ffn_down_shexp.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight q5_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down_exps.weight q5_K +blk.12.ffn_down_shexp.weight q5_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down_shexp.weight q5_K +blk.13.ffn_down_exps.weight q5_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down_shexp.weight q5_K +blk.14.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down_shexp.weight q5_K +blk.16.ffn_down_exps.weight q5_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down_exps.weight q5_K +blk.17.ffn_down_shexp.weight q5_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down_shexp.weight q5_K +blk.18.ffn_down_exps.weight q5_K +blk.19.ffn_down_shexp.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight q5_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down_shexp.weight q5_K +blk.20.ffn_down_exps.weight q5_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down_shexp.weight q5_K +blk.21.ffn_down_exps.weight q5_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down_shexp.weight q5_K +blk.22.ffn_down_exps.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.23.ffn_down_shexp.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down_shexp.weight q5_K +blk.24.ffn_down_exps.weight q5_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down_shexp.weight q5_K +blk.25.ffn_down_exps.weight q5_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down_shexp.weight q5_K +blk.26.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down_shexp.weight q5_K +blk.28.ffn_down_exps.weight q5_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down_exps.weight q5_K +blk.29.ffn_down_shexp.weight q5_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down_shexp.weight q5_K +blk.30.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down_shexp.weight q5_K +blk.32.ffn_down_exps.weight q5_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down_shexp.weight q5_K +blk.33.ffn_down_exps.weight q5_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down_exps.weight q5_K +blk.34.ffn_down_shexp.weight q5_K +blk.35.ffn_down_shexp.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight q5_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down_shexp.weight q5_K +blk.36.ffn_down_exps.weight q5_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down_shexp.weight q5_K +blk.37.ffn_down_exps.weight q5_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down_shexp.weight q5_K +blk.38.ffn_down_exps.weight q5_K +blk.39.ffn_down_shexp.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down_exps.weight q5_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down_exps.weight q5_K +blk.40.ffn_down_shexp.weight q5_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down_shexp.weight q5_K +blk.41.ffn_down_exps.weight q5_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down_shexp.weight q5_K +blk.42.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down_exps.weight q5_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down_shexp.weight q5_K +blk.44.ffn_down_exps.weight q5_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down_shexp.weight q5_K +blk.45.ffn_down_exps.weight q5_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down_exps.weight q5_K +blk.46.ffn_down_shexp.weight q5_K +blk.47.ffn_down_shexp.weight q5_K +blk.47.attn_output.weight q5_K +blk.47.attn_v.weight q5_K +output.weight q6_K +blk.47.ffn_down_exps.weight q5_K + +[Q4_K_S] q4_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +output.weight q6_K + +[Q4_K_M] q4_K +blk.0.attn_qkv.weight q5_K +blk.0.ffn_down_exps.weight q6_K +blk.0.ffn_down_shexp.weight q6_K +blk.1.attn_qkv.weight q5_K +blk.1.ffn_down_shexp.weight q6_K +blk.1.ffn_down_exps.weight q6_K +blk.2.attn_qkv.weight q5_K +blk.2.ffn_down_shexp.weight q6_K +blk.2.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.4.attn_qkv.weight q5_K +blk.4.ffn_down_shexp.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.5.attn_qkv.weight q5_K +blk.5.ffn_down_shexp.weight q6_K +blk.5.ffn_down_exps.weight q6_K +blk.6.attn_qkv.weight q5_K +blk.7.attn_v.weight q6_K +blk.8.attn_qkv.weight q5_K +blk.8.ffn_down_shexp.weight q6_K +blk.8.ffn_down_exps.weight q6_K +blk.9.attn_qkv.weight q5_K +blk.10.attn_qkv.weight q5_K +blk.11.ffn_down_shexp.weight q6_K +blk.11.attn_v.weight q6_K +blk.11.ffn_down_exps.weight q6_K +blk.12.attn_qkv.weight q5_K +blk.13.attn_qkv.weight q5_K +blk.14.attn_qkv.weight q5_K +blk.14.ffn_down_shexp.weight q6_K +blk.14.ffn_down_exps.weight q6_K +blk.15.attn_v.weight q6_K +blk.16.attn_qkv.weight q5_K +blk.17.attn_qkv.weight q5_K +blk.17.ffn_down_exps.weight q6_K +blk.17.ffn_down_shexp.weight q6_K +blk.18.attn_qkv.weight q5_K +blk.19.attn_v.weight q6_K +blk.20.attn_qkv.weight q5_K +blk.20.ffn_down_shexp.weight q6_K +blk.20.ffn_down_exps.weight q6_K +blk.21.attn_qkv.weight q5_K +blk.22.attn_qkv.weight q5_K +blk.23.ffn_down_exps.weight q6_K +blk.23.ffn_down_shexp.weight q6_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q5_K +blk.25.attn_qkv.weight q5_K +blk.26.attn_qkv.weight q5_K +blk.26.ffn_down_shexp.weight q6_K +blk.26.ffn_down_exps.weight q6_K +blk.28.attn_qkv.weight q5_K +blk.29.attn_qkv.weight q5_K +blk.29.ffn_down_exps.weight q6_K +blk.29.ffn_down_shexp.weight q6_K +blk.30.attn_qkv.weight q5_K +blk.32.attn_qkv.weight q5_K +blk.32.ffn_down_shexp.weight q6_K +blk.32.ffn_down_exps.weight q6_K +blk.33.attn_qkv.weight q5_K +blk.34.attn_qkv.weight q5_K +blk.35.ffn_down_shexp.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down_exps.weight q6_K +blk.36.attn_qkv.weight q5_K +blk.37.attn_qkv.weight q5_K +blk.38.attn_qkv.weight q5_K +blk.38.ffn_down_shexp.weight q6_K +blk.38.ffn_down_exps.weight q6_K +blk.40.attn_qkv.weight q5_K +blk.41.attn_qkv.weight q5_K +blk.41.ffn_down_shexp.weight q6_K +blk.41.ffn_down_exps.weight q6_K +blk.42.attn_qkv.weight q5_K +blk.42.ffn_down_shexp.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K +blk.43.ffn_down_exps.weight q6_K +blk.44.attn_qkv.weight q5_K +blk.44.ffn_down_shexp.weight q6_K +blk.44.ffn_down_exps.weight q6_K +blk.45.attn_qkv.weight q5_K +blk.45.ffn_down_shexp.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.46.attn_qkv.weight q5_K +blk.46.ffn_down_exps.weight q6_K +blk.46.ffn_down_shexp.weight q6_K +blk.47.ffn_down_shexp.weight q6_K +blk.47.attn_v.weight q6_K +output.weight q6_K +blk.47.ffn_down_exps.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +blk.0.attn_qkv.weight q6_K +blk.0.ffn_down_exps.weight q6_K +blk.0.ffn_down_shexp.weight q6_K +blk.1.attn_qkv.weight q6_K +blk.1.ffn_down_shexp.weight q6_K +blk.1.ffn_down_exps.weight q6_K +blk.2.attn_qkv.weight q6_K +blk.2.ffn_down_shexp.weight q6_K +blk.2.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.4.attn_qkv.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.5.attn_qkv.weight q6_K +blk.5.ffn_down_shexp.weight q6_K +blk.5.ffn_down_exps.weight q6_K +blk.6.attn_qkv.weight q6_K +blk.7.attn_v.weight q6_K +blk.8.attn_qkv.weight q6_K +blk.8.ffn_down_shexp.weight q6_K +blk.8.ffn_down_exps.weight q6_K +blk.9.attn_qkv.weight q6_K +blk.10.attn_qkv.weight q6_K +blk.11.ffn_down_shexp.weight q6_K +blk.11.attn_v.weight q6_K +blk.11.ffn_down_exps.weight q6_K +blk.12.attn_qkv.weight q6_K +blk.13.attn_qkv.weight q6_K +blk.14.attn_qkv.weight q6_K +blk.14.ffn_down_shexp.weight q6_K +blk.14.ffn_down_exps.weight q6_K +blk.15.attn_v.weight q6_K +blk.16.attn_qkv.weight q6_K +blk.17.attn_qkv.weight q6_K +blk.17.ffn_down_exps.weight q6_K +blk.17.ffn_down_shexp.weight q6_K +blk.18.attn_qkv.weight q6_K +blk.19.attn_v.weight q6_K +blk.20.attn_qkv.weight q6_K +blk.20.ffn_down_shexp.weight q6_K +blk.20.ffn_down_exps.weight q6_K +blk.21.attn_qkv.weight q6_K +blk.22.attn_qkv.weight q6_K +blk.23.ffn_down_exps.weight q6_K +blk.23.ffn_down_shexp.weight q6_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q6_K +blk.25.attn_qkv.weight q6_K +blk.26.attn_qkv.weight q6_K +blk.26.ffn_down_shexp.weight q6_K +blk.26.ffn_down_exps.weight q6_K +blk.28.attn_qkv.weight q6_K +blk.29.attn_qkv.weight q6_K +blk.29.ffn_down_exps.weight q6_K +blk.29.ffn_down_shexp.weight q6_K +blk.30.attn_qkv.weight q6_K +blk.32.attn_qkv.weight q6_K +blk.32.ffn_down_shexp.weight q6_K +blk.32.ffn_down_exps.weight q6_K +blk.33.attn_qkv.weight q6_K +blk.34.attn_qkv.weight q6_K +blk.35.ffn_down_shexp.weight q6_K +blk.35.attn_v.weight q6_K +blk.35.ffn_down_exps.weight q6_K +blk.36.attn_qkv.weight q6_K +blk.37.attn_qkv.weight q6_K +blk.38.attn_qkv.weight q6_K +blk.38.ffn_down_shexp.weight q6_K +blk.38.ffn_down_exps.weight q6_K +blk.40.attn_qkv.weight q6_K +blk.41.attn_qkv.weight q6_K +blk.41.ffn_down_shexp.weight q6_K +blk.41.ffn_down_exps.weight q6_K +blk.42.attn_qkv.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K +blk.43.ffn_down_exps.weight q6_K +blk.44.attn_qkv.weight q6_K +blk.44.ffn_down_shexp.weight q6_K +blk.44.ffn_down_exps.weight q6_K +blk.45.attn_qkv.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.46.attn_qkv.weight q6_K +blk.46.ffn_down_exps.weight q6_K +blk.46.ffn_down_shexp.weight q6_K +blk.47.ffn_down_shexp.weight q6_K +blk.47.attn_v.weight q6_K +output.weight q6_K +blk.47.ffn_down_exps.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q5_K + +[IQ2_XS] iq2_xs +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q5_K + +[Q2_K_S] q2_K +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q6_K + +[IQ3_XS] iq3_s +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.6.ffn_gate_exps.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs +blk.6.ffn_gate_shexp.weight iq3_xxs +blk.6.ffn_up_shexp.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_shexp.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.24.ffn_gate_shexp.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs +blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_up_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs +blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_up_exps.weight iq3_xxs +blk.34.ffn_gate_exps.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K +blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_up_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs +blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs +blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_up_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.39.ffn_gate_shexp.weight iq3_xxs +blk.39.ffn_up_shexp.weight iq3_xxs +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q4_K +blk.39.ffn_gate_exps.weight iq3_xxs +blk.39.ffn_up_exps.weight iq3_xxs +blk.40.ffn_gate_exps.weight iq3_xxs +blk.40.ffn_up_exps.weight iq3_xxs +blk.40.ffn_gate_shexp.weight iq3_xxs +blk.40.ffn_up_shexp.weight iq3_xxs +blk.41.ffn_gate_shexp.weight iq3_xxs +blk.41.ffn_up_shexp.weight iq3_xxs +blk.41.ffn_gate_exps.weight iq3_xxs +blk.41.ffn_up_exps.weight iq3_xxs +blk.43.attn_k.weight iq3_xxs +blk.43.attn_q.weight iq3_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_k.weight iq3_xxs +blk.47.attn_q.weight iq3_xxs +blk.47.attn_v.weight q4_K +output.weight q6_K + +[IQ3_XXS] iq3_xxs +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.9.ffn_down_exps.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.14.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.16.ffn_down_exps.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.18.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.21.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.24.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.25.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.26.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.28.ffn_down_exps.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.30.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.32.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.33.ffn_down_exps.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.37.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.41.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.42.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.43.attn_k.weight iq2_s +blk.43.attn_output.weight iq3_s +blk.43.attn_q.weight iq2_s +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.44.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.47.attn_k.weight iq2_s +blk.47.attn_output.weight iq3_s +blk.47.attn_q.weight iq2_s +blk.47.attn_v.weight q4_K +output.weight q5_K +blk.47.ffn_down_exps.weight q3_K + +[IQ1_S] iq1_s +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +output.weight q5_K + +[IQ4_NL] iq4_nl +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +output.weight q6_K + +[IQ3_S] iq3_s +blk.3.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q6_K + +[IQ3_M] iq3_s +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +output.weight q6_K + +[IQ2_S] iq2_xs +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight iq3_s +blk.0.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s +blk.2.ffn_down_exps.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +output.weight q5_K + +[IQ2_M] iq2_s +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight iq3_s +blk.0.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s +blk.2.ffn_down_exps.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +output.weight q5_K + +[IQ4_XS] iq4_xs +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +output.weight q6_K + +[IQ1_M] iq1_m +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +output.weight q5_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +token_embd.weight q4_K +output.weight q6_K + +[TQ2_0] tq2_0 +token_embd.weight q4_K +output.weight q6_K + +[MXFP4_MOE] mxfp4 +token_embd.weight q8_0 +blk.0.ssm_ba.weight q8_0 +blk.0.attn_qkv.weight q8_0 +blk.0.attn_gate.weight q8_0 +blk.0.ssm_out.weight q8_0 +blk.0.ffn_down_shexp.weight q8_0 +blk.0.ffn_gate_shexp.weight q8_0 +blk.0.ffn_up_shexp.weight q8_0 +blk.1.ssm_ba.weight q8_0 +blk.1.attn_qkv.weight q8_0 +blk.1.attn_gate.weight q8_0 +blk.1.ssm_out.weight q8_0 +blk.1.ffn_down_shexp.weight q8_0 +blk.1.ffn_gate_shexp.weight q8_0 +blk.1.ffn_up_shexp.weight q8_0 +blk.2.ssm_ba.weight q8_0 +blk.2.attn_qkv.weight q8_0 +blk.2.attn_gate.weight q8_0 +blk.2.ssm_out.weight q8_0 +blk.2.ffn_down_shexp.weight q8_0 +blk.2.ffn_gate_shexp.weight q8_0 +blk.2.ffn_up_shexp.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.4.ssm_ba.weight q8_0 +blk.4.attn_qkv.weight q8_0 +blk.4.attn_gate.weight q8_0 +blk.4.ssm_out.weight q8_0 +blk.4.ffn_down_shexp.weight q8_0 +blk.4.ffn_gate_shexp.weight q8_0 +blk.4.ffn_up_shexp.weight q8_0 +blk.5.ssm_ba.weight q8_0 +blk.5.attn_qkv.weight q8_0 +blk.5.attn_gate.weight q8_0 +blk.5.ssm_out.weight q8_0 +blk.5.ffn_down_shexp.weight q8_0 +blk.5.ffn_gate_shexp.weight q8_0 +blk.5.ffn_up_shexp.weight q8_0 +blk.6.ssm_ba.weight q8_0 +blk.6.attn_qkv.weight q8_0 +blk.6.attn_gate.weight q8_0 +blk.6.ssm_out.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_gate_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.8.ssm_ba.weight q8_0 +blk.8.attn_qkv.weight q8_0 +blk.8.attn_gate.weight q8_0 +blk.8.ssm_out.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_gate_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 +blk.9.ssm_ba.weight q8_0 +blk.9.attn_qkv.weight q8_0 +blk.9.attn_gate.weight q8_0 +blk.9.ssm_out.weight q8_0 +blk.9.ffn_down_shexp.weight q8_0 +blk.9.ffn_gate_shexp.weight q8_0 +blk.9.ffn_up_shexp.weight q8_0 +blk.10.ssm_ba.weight q8_0 +blk.10.attn_qkv.weight q8_0 +blk.10.attn_gate.weight q8_0 +blk.10.ssm_out.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_gate_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.12.ssm_ba.weight q8_0 +blk.12.attn_qkv.weight q8_0 +blk.12.attn_gate.weight q8_0 +blk.12.ssm_out.weight q8_0 +blk.12.ffn_down_shexp.weight q8_0 +blk.12.ffn_gate_shexp.weight q8_0 +blk.12.ffn_up_shexp.weight q8_0 +blk.13.ssm_ba.weight q8_0 +blk.13.attn_qkv.weight q8_0 +blk.13.attn_gate.weight q8_0 +blk.13.ssm_out.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_gate_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 +blk.14.ssm_ba.weight q8_0 +blk.14.attn_qkv.weight q8_0 +blk.14.attn_gate.weight q8_0 +blk.14.ssm_out.weight q8_0 +blk.14.ffn_down_shexp.weight q8_0 +blk.14.ffn_gate_shexp.weight q8_0 +blk.14.ffn_up_shexp.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.16.ssm_ba.weight q8_0 +blk.16.attn_qkv.weight q8_0 +blk.16.attn_gate.weight q8_0 +blk.16.ssm_out.weight q8_0 +blk.16.ffn_down_shexp.weight q8_0 +blk.16.ffn_gate_shexp.weight q8_0 +blk.16.ffn_up_shexp.weight q8_0 +blk.17.ssm_ba.weight q8_0 +blk.17.attn_qkv.weight q8_0 +blk.17.attn_gate.weight q8_0 +blk.17.ssm_out.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_gate_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 +blk.18.ssm_ba.weight q8_0 +blk.18.attn_qkv.weight q8_0 +blk.18.attn_gate.weight q8_0 +blk.18.ssm_out.weight q8_0 +blk.18.ffn_down_shexp.weight q8_0 +blk.18.ffn_gate_shexp.weight q8_0 +blk.18.ffn_up_shexp.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.20.ssm_ba.weight q8_0 +blk.20.attn_qkv.weight q8_0 +blk.20.attn_gate.weight q8_0 +blk.20.ssm_out.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_gate_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 +blk.21.ssm_ba.weight q8_0 +blk.21.attn_qkv.weight q8_0 +blk.21.attn_gate.weight q8_0 +blk.21.ssm_out.weight q8_0 +blk.21.ffn_down_shexp.weight q8_0 +blk.21.ffn_gate_shexp.weight q8_0 +blk.21.ffn_up_shexp.weight q8_0 +blk.22.ssm_ba.weight q8_0 +blk.22.attn_qkv.weight q8_0 +blk.22.attn_gate.weight q8_0 +blk.22.ssm_out.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_gate_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.24.ssm_ba.weight q8_0 +blk.24.attn_qkv.weight q8_0 +blk.24.attn_gate.weight q8_0 +blk.24.ssm_out.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_gate_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 +blk.25.ssm_ba.weight q8_0 +blk.25.attn_qkv.weight q8_0 +blk.25.attn_gate.weight q8_0 +blk.25.ssm_out.weight q8_0 +blk.25.ffn_down_shexp.weight q8_0 +blk.25.ffn_gate_shexp.weight q8_0 +blk.25.ffn_up_shexp.weight q8_0 +blk.26.ssm_ba.weight q8_0 +blk.26.attn_qkv.weight q8_0 +blk.26.attn_gate.weight q8_0 +blk.26.ssm_out.weight q8_0 +blk.26.ffn_down_shexp.weight q8_0 +blk.26.ffn_gate_shexp.weight q8_0 +blk.26.ffn_up_shexp.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.28.ssm_ba.weight q8_0 +blk.28.attn_qkv.weight q8_0 +blk.28.attn_gate.weight q8_0 +blk.28.ssm_out.weight q8_0 +blk.28.ffn_down_shexp.weight q8_0 +blk.28.ffn_gate_shexp.weight q8_0 +blk.28.ffn_up_shexp.weight q8_0 +blk.29.ssm_ba.weight q8_0 +blk.29.attn_qkv.weight q8_0 +blk.29.attn_gate.weight q8_0 +blk.29.ssm_out.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_gate_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 +blk.30.ssm_ba.weight q8_0 +blk.30.attn_qkv.weight q8_0 +blk.30.attn_gate.weight q8_0 +blk.30.ssm_out.weight q8_0 +blk.30.ffn_down_shexp.weight q8_0 +blk.30.ffn_gate_shexp.weight q8_0 +blk.30.ffn_up_shexp.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.32.ssm_ba.weight q8_0 +blk.32.attn_qkv.weight q8_0 +blk.32.attn_gate.weight q8_0 +blk.32.ssm_out.weight q8_0 +blk.32.ffn_down_shexp.weight q8_0 +blk.32.ffn_gate_shexp.weight q8_0 +blk.32.ffn_up_shexp.weight q8_0 +blk.33.ssm_ba.weight q8_0 +blk.33.attn_qkv.weight q8_0 +blk.33.attn_gate.weight q8_0 +blk.33.ssm_out.weight q8_0 +blk.33.ffn_down_shexp.weight q8_0 +blk.33.ffn_gate_shexp.weight q8_0 +blk.33.ffn_up_shexp.weight q8_0 +blk.34.ssm_ba.weight q8_0 +blk.34.attn_qkv.weight q8_0 +blk.34.attn_gate.weight q8_0 +blk.34.ssm_out.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_gate_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.36.ssm_ba.weight q8_0 +blk.36.attn_qkv.weight q8_0 +blk.36.attn_gate.weight q8_0 +blk.36.ssm_out.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_gate_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 +blk.37.ssm_ba.weight q8_0 +blk.37.attn_qkv.weight q8_0 +blk.37.attn_gate.weight q8_0 +blk.37.ssm_out.weight q8_0 +blk.37.ffn_down_shexp.weight q8_0 +blk.37.ffn_gate_shexp.weight q8_0 +blk.37.ffn_up_shexp.weight q8_0 +blk.38.ssm_ba.weight q8_0 +blk.38.attn_qkv.weight q8_0 +blk.38.attn_gate.weight q8_0 +blk.38.ssm_out.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_gate_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.40.ssm_ba.weight q8_0 +blk.40.attn_qkv.weight q8_0 +blk.40.attn_gate.weight q8_0 +blk.40.ssm_out.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_gate_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 +blk.41.ssm_ba.weight q8_0 +blk.41.attn_qkv.weight q8_0 +blk.41.attn_gate.weight q8_0 +blk.41.ssm_out.weight q8_0 +blk.41.ffn_down_shexp.weight q8_0 +blk.41.ffn_gate_shexp.weight q8_0 +blk.41.ffn_up_shexp.weight q8_0 +blk.42.ssm_ba.weight q8_0 +blk.42.attn_qkv.weight q8_0 +blk.42.attn_gate.weight q8_0 +blk.42.ssm_out.weight q8_0 +blk.42.ffn_down_shexp.weight q8_0 +blk.42.ffn_gate_shexp.weight q8_0 +blk.42.ffn_up_shexp.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 +blk.43.attn_k.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q.weight q8_0 +blk.43.attn_v.weight q8_0 +blk.44.ssm_ba.weight q8_0 +blk.44.attn_qkv.weight q8_0 +blk.44.attn_gate.weight q8_0 +blk.44.ssm_out.weight q8_0 +blk.44.ffn_down_shexp.weight q8_0 +blk.44.ffn_gate_shexp.weight q8_0 +blk.44.ffn_up_shexp.weight q8_0 +blk.45.ssm_ba.weight q8_0 +blk.45.attn_qkv.weight q8_0 +blk.45.attn_gate.weight q8_0 +blk.45.ssm_out.weight q8_0 +blk.45.ffn_down_shexp.weight q8_0 +blk.45.ffn_gate_shexp.weight q8_0 +blk.45.ffn_up_shexp.weight q8_0 +blk.46.ssm_ba.weight q8_0 +blk.46.attn_qkv.weight q8_0 +blk.46.attn_gate.weight q8_0 +blk.46.ssm_out.weight q8_0 +blk.46.ffn_down_shexp.weight q8_0 +blk.46.ffn_gate_shexp.weight q8_0 +blk.46.ffn_up_shexp.weight q8_0 +blk.47.ffn_down_shexp.weight q8_0 +blk.47.ffn_gate_shexp.weight q8_0 +blk.47.ffn_up_shexp.weight q8_0 +blk.47.attn_k.weight q8_0 +blk.47.attn_output.weight q8_0 +blk.47.attn_q.weight q8_0 +blk.47.attn_v.weight q8_0 +output.weight q8_0 diff --git a/tests/snapshots/qwen3.5-27b.schema b/tests/snapshots/qwen3.5-27b.schema new file mode 100644 index 0000000000..f1f655a847 --- /dev/null +++ b/tests/snapshots/qwen3.5-27b.schema @@ -0,0 +1,1837 @@ +# Model: Qwen3.5-27B +# n_embd=5120, n_ff=17408, n_vocab=248320, n_layer=64, n_head=24, n_head_kv=4 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q6_K +blk.0.ffn_down.weight q3_K +blk.1.ffn_down.weight q3_K +blk.2.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q3_K +blk.4.ffn_down.weight q3_K +blk.5.ffn_down.weight q3_K +blk.6.ffn_down.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q3_K +blk.8.ffn_down.weight q3_K +blk.9.ffn_down.weight q3_K +blk.10.ffn_down.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q3_K +blk.12.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K +blk.14.ffn_down.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K +blk.17.ffn_down.weight q3_K +blk.18.ffn_down.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q3_K +blk.20.ffn_down.weight q3_K +blk.21.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K +blk.25.ffn_down.weight q3_K +blk.26.ffn_down.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q3_K +blk.28.ffn_down.weight q3_K +blk.29.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q3_K +blk.32.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K +blk.34.ffn_down.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q3_K +blk.36.ffn_down.weight q3_K +blk.37.ffn_down.weight q3_K +blk.38.ffn_down.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q3_K +blk.40.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K +blk.42.ffn_down.weight q3_K +blk.43.attn_output.weight q3_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K +blk.45.ffn_down.weight q3_K +blk.46.ffn_down.weight q3_K +blk.47.attn_output.weight q3_K +blk.47.attn_v.weight q4_K +blk.47.ffn_down.weight q3_K +blk.48.ffn_down.weight q3_K +blk.49.ffn_down.weight q3_K +blk.50.ffn_down.weight q3_K +blk.51.attn_output.weight q3_K +blk.51.attn_v.weight q4_K +blk.51.ffn_down.weight q3_K +blk.52.ffn_down.weight q3_K +blk.53.ffn_down.weight q3_K +blk.54.ffn_down.weight q3_K +blk.55.attn_output.weight q3_K +blk.55.attn_v.weight q4_K +blk.55.ffn_down.weight q3_K +blk.56.ffn_down.weight q3_K +blk.57.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K +blk.59.attn_output.weight q3_K +blk.59.attn_v.weight q4_K +blk.59.ffn_down.weight q3_K +blk.60.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K +blk.62.ffn_down.weight q3_K +blk.63.attn_output.weight q3_K +blk.63.attn_v.weight q4_K +blk.63.ffn_down.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +blk.47.ffn_down.weight q4_K +blk.48.attn_qkv.weight q4_K +blk.48.ffn_down.weight q4_K +blk.49.attn_qkv.weight q4_K +blk.49.ffn_down.weight q4_K +blk.50.attn_qkv.weight q4_K +blk.50.ffn_down.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q4_K +blk.51.ffn_down.weight q4_K +blk.52.attn_qkv.weight q4_K +blk.52.ffn_down.weight q4_K +blk.53.attn_qkv.weight q4_K +blk.53.ffn_down.weight q4_K +blk.54.attn_qkv.weight q4_K +blk.54.ffn_down.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q4_K +blk.55.ffn_down.weight q4_K +blk.56.attn_qkv.weight q4_K +blk.56.ffn_down.weight q4_K +blk.57.attn_qkv.weight q4_K +blk.57.ffn_down.weight q4_K +blk.58.attn_qkv.weight q4_K +blk.58.ffn_down.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q4_K +blk.59.ffn_down.weight q4_K +blk.60.attn_qkv.weight q4_K +blk.60.ffn_down.weight q4_K +blk.61.attn_qkv.weight q4_K +blk.61.ffn_down.weight q4_K +blk.62.attn_qkv.weight q4_K +blk.62.ffn_down.weight q4_K +blk.63.attn_output.weight q4_K +blk.63.attn_v.weight q4_K +blk.63.ffn_down.weight q4_K + +[Q3_K_L] q3_K +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down.weight q5_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down.weight q5_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down.weight q5_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down.weight q5_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down.weight q5_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down.weight q5_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down.weight q5_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down.weight q5_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down.weight q5_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down.weight q5_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down.weight q5_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down.weight q5_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down.weight q5_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down.weight q5_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down.weight q5_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down.weight q5_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down.weight q5_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down.weight q5_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q5_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down.weight q5_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down.weight q5_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down.weight q5_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down.weight q5_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down.weight q5_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q5_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down.weight q5_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down.weight q5_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down.weight q5_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down.weight q5_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down.weight q5_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down.weight q5_K +blk.47.attn_output.weight q5_K +blk.47.attn_v.weight q5_K +blk.47.ffn_down.weight q5_K +blk.48.attn_qkv.weight q4_K +blk.48.ffn_down.weight q5_K +blk.49.attn_qkv.weight q4_K +blk.49.ffn_down.weight q5_K +blk.50.attn_qkv.weight q4_K +blk.50.ffn_down.weight q5_K +blk.51.attn_output.weight q5_K +blk.51.attn_v.weight q5_K +blk.51.ffn_down.weight q5_K +blk.52.attn_qkv.weight q4_K +blk.52.ffn_down.weight q5_K +blk.53.attn_qkv.weight q4_K +blk.53.ffn_down.weight q5_K +blk.54.attn_qkv.weight q4_K +blk.54.ffn_down.weight q5_K +blk.55.attn_output.weight q5_K +blk.55.attn_v.weight q5_K +blk.55.ffn_down.weight q5_K +blk.56.attn_qkv.weight q4_K +blk.56.ffn_down.weight q5_K +blk.57.attn_qkv.weight q4_K +blk.57.ffn_down.weight q5_K +blk.58.attn_qkv.weight q4_K +blk.58.ffn_down.weight q5_K +blk.59.attn_output.weight q5_K +blk.59.attn_v.weight q5_K +blk.59.ffn_down.weight q5_K +blk.60.attn_qkv.weight q4_K +blk.60.ffn_down.weight q5_K +blk.61.attn_qkv.weight q4_K +blk.61.ffn_down.weight q5_K +blk.62.attn_qkv.weight q4_K +blk.62.ffn_down.weight q5_K +blk.63.attn_output.weight q5_K +blk.63.attn_v.weight q5_K +blk.63.ffn_down.weight q5_K + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.6.ffn_down.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K + +[Q4_K_M] q4_K +output.weight q6_K +blk.0.attn_qkv.weight q5_K +blk.0.ffn_down.weight q6_K +blk.1.attn_qkv.weight q5_K +blk.1.ffn_down.weight q6_K +blk.2.attn_qkv.weight q5_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.4.attn_qkv.weight q5_K +blk.4.ffn_down.weight q6_K +blk.5.attn_qkv.weight q5_K +blk.5.ffn_down.weight q6_K +blk.6.attn_qkv.weight q5_K +blk.6.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.8.attn_qkv.weight q5_K +blk.9.attn_qkv.weight q5_K +blk.10.attn_qkv.weight q5_K +blk.10.ffn_down.weight q6_K +blk.11.attn_v.weight q6_K +blk.12.attn_qkv.weight q5_K +blk.13.attn_qkv.weight q5_K +blk.13.ffn_down.weight q6_K +blk.14.attn_qkv.weight q5_K +blk.15.attn_v.weight q6_K +blk.16.attn_qkv.weight q5_K +blk.16.ffn_down.weight q6_K +blk.17.attn_qkv.weight q5_K +blk.18.attn_qkv.weight q5_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down.weight q6_K +blk.20.attn_qkv.weight q5_K +blk.21.attn_qkv.weight q5_K +blk.22.attn_qkv.weight q5_K +blk.22.ffn_down.weight q6_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q5_K +blk.25.attn_qkv.weight q5_K +blk.25.ffn_down.weight q6_K +blk.26.attn_qkv.weight q5_K +blk.27.attn_v.weight q6_K +blk.28.attn_qkv.weight q5_K +blk.28.ffn_down.weight q6_K +blk.29.attn_qkv.weight q5_K +blk.30.attn_qkv.weight q5_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.32.attn_qkv.weight q5_K +blk.33.attn_qkv.weight q5_K +blk.34.attn_qkv.weight q5_K +blk.34.ffn_down.weight q6_K +blk.36.attn_qkv.weight q5_K +blk.37.attn_qkv.weight q5_K +blk.37.ffn_down.weight q6_K +blk.38.attn_qkv.weight q5_K +blk.40.attn_qkv.weight q5_K +blk.40.ffn_down.weight q6_K +blk.41.attn_qkv.weight q5_K +blk.42.attn_qkv.weight q5_K +blk.43.attn_v.weight q6_K +blk.43.ffn_down.weight q6_K +blk.44.attn_qkv.weight q5_K +blk.45.attn_qkv.weight q5_K +blk.46.attn_qkv.weight q5_K +blk.46.ffn_down.weight q6_K +blk.48.attn_qkv.weight q5_K +blk.49.attn_qkv.weight q5_K +blk.49.ffn_down.weight q6_K +blk.50.attn_qkv.weight q5_K +blk.52.attn_qkv.weight q5_K +blk.52.ffn_down.weight q6_K +blk.53.attn_qkv.weight q5_K +blk.54.attn_qkv.weight q5_K +blk.55.attn_v.weight q6_K +blk.55.ffn_down.weight q6_K +blk.56.attn_qkv.weight q5_K +blk.56.ffn_down.weight q6_K +blk.57.attn_qkv.weight q5_K +blk.57.ffn_down.weight q6_K +blk.58.attn_qkv.weight q5_K +blk.58.ffn_down.weight q6_K +blk.59.ffn_down.weight q6_K +blk.60.attn_qkv.weight q5_K +blk.60.ffn_down.weight q6_K +blk.61.attn_qkv.weight q5_K +blk.61.ffn_down.weight q6_K +blk.62.attn_qkv.weight q5_K +blk.62.ffn_down.weight q6_K +blk.63.ffn_down.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +output.weight q6_K +blk.0.attn_qkv.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.attn_qkv.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_qkv.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.4.attn_qkv.weight q6_K +blk.4.ffn_down.weight q6_K +blk.5.attn_qkv.weight q6_K +blk.5.ffn_down.weight q6_K +blk.6.attn_qkv.weight q6_K +blk.6.ffn_down.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down.weight q6_K +blk.8.attn_qkv.weight q6_K +blk.9.attn_qkv.weight q6_K +blk.10.attn_qkv.weight q6_K +blk.10.ffn_down.weight q6_K +blk.11.attn_v.weight q6_K +blk.12.attn_qkv.weight q6_K +blk.13.attn_qkv.weight q6_K +blk.13.ffn_down.weight q6_K +blk.14.attn_qkv.weight q6_K +blk.15.attn_v.weight q6_K +blk.16.attn_qkv.weight q6_K +blk.16.ffn_down.weight q6_K +blk.17.attn_qkv.weight q6_K +blk.18.attn_qkv.weight q6_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down.weight q6_K +blk.20.attn_qkv.weight q6_K +blk.21.attn_qkv.weight q6_K +blk.22.attn_qkv.weight q6_K +blk.22.ffn_down.weight q6_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q6_K +blk.25.attn_qkv.weight q6_K +blk.25.ffn_down.weight q6_K +blk.26.attn_qkv.weight q6_K +blk.27.attn_v.weight q6_K +blk.28.attn_qkv.weight q6_K +blk.28.ffn_down.weight q6_K +blk.29.attn_qkv.weight q6_K +blk.30.attn_qkv.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down.weight q6_K +blk.32.attn_qkv.weight q6_K +blk.33.attn_qkv.weight q6_K +blk.34.attn_qkv.weight q6_K +blk.34.ffn_down.weight q6_K +blk.36.attn_qkv.weight q6_K +blk.37.attn_qkv.weight q6_K +blk.37.ffn_down.weight q6_K +blk.38.attn_qkv.weight q6_K +blk.40.attn_qkv.weight q6_K +blk.40.ffn_down.weight q6_K +blk.41.attn_qkv.weight q6_K +blk.42.attn_qkv.weight q6_K +blk.43.attn_v.weight q6_K +blk.43.ffn_down.weight q6_K +blk.44.attn_qkv.weight q6_K +blk.45.attn_qkv.weight q6_K +blk.46.attn_qkv.weight q6_K +blk.46.ffn_down.weight q6_K +blk.48.attn_qkv.weight q6_K +blk.49.attn_qkv.weight q6_K +blk.49.ffn_down.weight q6_K +blk.50.attn_qkv.weight q6_K +blk.52.attn_qkv.weight q6_K +blk.52.ffn_down.weight q6_K +blk.53.attn_qkv.weight q6_K +blk.54.attn_qkv.weight q6_K +blk.55.attn_v.weight q6_K +blk.55.ffn_down.weight q6_K +blk.56.attn_qkv.weight q6_K +blk.56.ffn_down.weight q6_K +blk.57.attn_qkv.weight q6_K +blk.57.ffn_down.weight q6_K +blk.58.attn_qkv.weight q6_K +blk.58.ffn_down.weight q6_K +blk.59.ffn_down.weight q6_K +blk.60.attn_qkv.weight q6_K +blk.60.ffn_down.weight q6_K +blk.61.attn_qkv.weight q6_K +blk.61.ffn_down.weight q6_K +blk.62.attn_qkv.weight q6_K +blk.62.ffn_down.weight q6_K +blk.63.ffn_down.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.6.ffn_down.weight q2_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.63.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.6.ffn_down.weight q2_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.63.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q6_K +blk.0.ffn_down.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K +blk.6.ffn_down.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.63.attn_v.weight q4_K + +[IQ3_XS] iq3_s +output.weight q6_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.8.ffn_gate.weight iq3_xxs +blk.8.ffn_up.weight iq3_xxs +blk.9.ffn_gate.weight iq3_xxs +blk.9.ffn_up.weight iq3_xxs +blk.10.ffn_gate.weight iq3_xxs +blk.10.ffn_up.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate.weight iq3_xxs +blk.11.ffn_up.weight iq3_xxs +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate.weight iq3_xxs +blk.15.ffn_up.weight iq3_xxs +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs +blk.18.ffn_gate.weight iq3_xxs +blk.18.ffn_up.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs +blk.21.ffn_gate.weight iq3_xxs +blk.21.ffn_up.weight iq3_xxs +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs +blk.29.ffn_gate.weight iq3_xxs +blk.29.ffn_up.weight iq3_xxs +blk.30.ffn_gate.weight iq3_xxs +blk.30.ffn_up.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_gate.weight iq3_xxs +blk.31.ffn_up.weight iq3_xxs +blk.32.ffn_gate.weight iq3_xxs +blk.32.ffn_up.weight iq3_xxs +blk.33.ffn_gate.weight iq3_xxs +blk.33.ffn_up.weight iq3_xxs +blk.34.ffn_gate.weight iq3_xxs +blk.34.ffn_up.weight iq3_xxs +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K +blk.35.ffn_gate.weight iq3_xxs +blk.35.ffn_up.weight iq3_xxs +blk.36.ffn_gate.weight iq3_xxs +blk.36.ffn_up.weight iq3_xxs +blk.37.ffn_gate.weight iq3_xxs +blk.37.ffn_up.weight iq3_xxs +blk.38.ffn_gate.weight iq3_xxs +blk.38.ffn_up.weight iq3_xxs +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q4_K +blk.39.ffn_gate.weight iq3_xxs +blk.39.ffn_up.weight iq3_xxs +blk.40.ffn_gate.weight iq3_xxs +blk.40.ffn_up.weight iq3_xxs +blk.41.ffn_gate.weight iq3_xxs +blk.41.ffn_up.weight iq3_xxs +blk.42.ffn_gate.weight iq3_xxs +blk.42.ffn_up.weight iq3_xxs +blk.43.attn_k.weight iq3_xxs +blk.43.attn_q.weight iq3_xxs +blk.43.attn_v.weight q4_K +blk.43.ffn_gate.weight iq3_xxs +blk.43.ffn_up.weight iq3_xxs +blk.44.ffn_gate.weight iq3_xxs +blk.44.ffn_up.weight iq3_xxs +blk.45.ffn_gate.weight iq3_xxs +blk.45.ffn_up.weight iq3_xxs +blk.46.ffn_gate.weight iq3_xxs +blk.46.ffn_up.weight iq3_xxs +blk.47.attn_k.weight iq3_xxs +blk.47.attn_q.weight iq3_xxs +blk.47.attn_v.weight q4_K +blk.47.ffn_gate.weight iq3_xxs +blk.47.ffn_up.weight iq3_xxs +blk.48.ffn_gate.weight iq3_xxs +blk.48.ffn_up.weight iq3_xxs +blk.49.ffn_gate.weight iq3_xxs +blk.49.ffn_up.weight iq3_xxs +blk.50.ffn_gate.weight iq3_xxs +blk.50.ffn_up.weight iq3_xxs +blk.51.attn_k.weight iq3_xxs +blk.51.attn_q.weight iq3_xxs +blk.51.attn_v.weight q4_K +blk.51.ffn_gate.weight iq3_xxs +blk.51.ffn_up.weight iq3_xxs +blk.52.ffn_gate.weight iq3_xxs +blk.52.ffn_up.weight iq3_xxs +blk.53.ffn_gate.weight iq3_xxs +blk.53.ffn_up.weight iq3_xxs +blk.54.ffn_gate.weight iq3_xxs +blk.54.ffn_up.weight iq3_xxs +blk.55.attn_k.weight iq3_xxs +blk.55.attn_q.weight iq3_xxs +blk.55.attn_v.weight q4_K +blk.55.ffn_gate.weight iq3_xxs +blk.55.ffn_up.weight iq3_xxs +blk.59.attn_k.weight iq3_xxs +blk.59.attn_q.weight iq3_xxs +blk.59.attn_v.weight q4_K +blk.63.attn_k.weight iq3_xxs +blk.63.attn_q.weight iq3_xxs +blk.63.attn_v.weight q4_K + +[IQ3_XXS] iq3_xxs +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K +blk.6.ffn_down.weight q4_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q4_K +blk.8.ffn_down.weight q3_K +blk.9.ffn_down.weight q3_K +blk.10.ffn_down.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q3_K +blk.12.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K +blk.14.ffn_down.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K +blk.17.ffn_down.weight q3_K +blk.18.ffn_down.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down.weight q3_K +blk.20.ffn_down.weight q3_K +blk.21.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.23.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K +blk.25.ffn_down.weight q3_K +blk.26.ffn_down.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down.weight q3_K +blk.28.ffn_down.weight q3_K +blk.29.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down.weight q3_K +blk.32.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K +blk.34.ffn_down.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down.weight q3_K +blk.36.ffn_down.weight q3_K +blk.37.ffn_down.weight q3_K +blk.38.ffn_down.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q4_K +blk.39.ffn_down.weight q3_K +blk.40.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K +blk.42.ffn_down.weight q3_K +blk.43.attn_k.weight iq2_s +blk.43.attn_output.weight iq3_s +blk.43.attn_q.weight iq2_s +blk.43.attn_v.weight q4_K +blk.43.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K +blk.45.ffn_down.weight q3_K +blk.46.ffn_down.weight q3_K +blk.47.attn_k.weight iq2_s +blk.47.attn_output.weight iq3_s +blk.47.attn_q.weight iq2_s +blk.47.attn_v.weight q4_K +blk.47.ffn_down.weight q3_K +blk.48.ffn_down.weight q3_K +blk.49.ffn_down.weight q3_K +blk.50.ffn_down.weight q3_K +blk.51.attn_k.weight iq2_s +blk.51.attn_output.weight iq3_s +blk.51.attn_q.weight iq2_s +blk.51.attn_v.weight q4_K +blk.51.ffn_down.weight q3_K +blk.52.ffn_down.weight q3_K +blk.53.ffn_down.weight q3_K +blk.54.ffn_down.weight q3_K +blk.55.attn_k.weight iq2_s +blk.55.attn_output.weight iq3_s +blk.55.attn_q.weight iq2_s +blk.55.attn_v.weight q4_K +blk.55.ffn_down.weight q3_K +blk.56.ffn_down.weight q3_K +blk.57.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K +blk.59.attn_k.weight iq2_s +blk.59.attn_output.weight iq3_s +blk.59.attn_q.weight iq2_s +blk.59.attn_v.weight q4_K +blk.59.ffn_down.weight q3_K +blk.60.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K +blk.62.ffn_down.weight q3_K +blk.63.attn_k.weight iq2_s +blk.63.attn_output.weight iq3_s +blk.63.attn_q.weight iq2_s +blk.63.attn_v.weight q4_K +blk.63.ffn_down.weight q3_K + +[IQ1_S] iq1_s +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.6.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K +blk.63.attn_output.weight iq2_xxs +blk.63.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.6.ffn_down.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.63.attn_v.weight q5_K + +[IQ3_S] iq3_s +output.weight q6_K +blk.3.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K +blk.63.attn_v.weight q4_K + +[IQ3_M] iq3_s +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +blk.48.attn_qkv.weight q4_K +blk.49.attn_qkv.weight q4_K +blk.50.attn_qkv.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q4_K +blk.52.attn_qkv.weight q4_K +blk.53.attn_qkv.weight q4_K +blk.54.attn_qkv.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q4_K +blk.56.attn_qkv.weight q4_K +blk.57.attn_qkv.weight q4_K +blk.58.attn_qkv.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q4_K +blk.60.attn_qkv.weight q4_K +blk.61.attn_qkv.weight q4_K +blk.62.attn_qkv.weight q4_K +blk.63.attn_output.weight q4_K +blk.63.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.4.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s +blk.6.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K +blk.63.attn_output.weight iq3_s +blk.63.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down.weight iq3_s +blk.1.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s +blk.4.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s +blk.6.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K +blk.63.attn_output.weight iq3_s +blk.63.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K +blk.4.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K +blk.6.ffn_down.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.59.attn_v.weight q5_K +blk.63.attn_v.weight q5_K + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down.weight q2_K +blk.1.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K +blk.4.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K +blk.6.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_down.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K +blk.63.attn_output.weight iq2_xxs +blk.63.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q6_K +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q6_K +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_gate.weight q8_0 +blk.0.attn_qkv.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.0.ssm_alpha.weight q8_0 +blk.0.ssm_beta.weight q8_0 +blk.0.ssm_out.weight q8_0 +blk.1.attn_gate.weight q8_0 +blk.1.attn_qkv.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.1.ssm_alpha.weight q8_0 +blk.1.ssm_beta.weight q8_0 +blk.1.ssm_out.weight q8_0 +blk.2.attn_gate.weight q8_0 +blk.2.attn_qkv.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.2.ssm_alpha.weight q8_0 +blk.2.ssm_beta.weight q8_0 +blk.2.ssm_out.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 +blk.4.attn_gate.weight q8_0 +blk.4.attn_qkv.weight q8_0 +blk.4.ffn_down.weight q8_0 +blk.4.ffn_gate.weight q8_0 +blk.4.ffn_up.weight q8_0 +blk.4.ssm_alpha.weight q8_0 +blk.4.ssm_beta.weight q8_0 +blk.4.ssm_out.weight q8_0 +blk.5.attn_gate.weight q8_0 +blk.5.attn_qkv.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 +blk.5.ssm_alpha.weight q8_0 +blk.5.ssm_beta.weight q8_0 +blk.5.ssm_out.weight q8_0 +blk.6.attn_gate.weight q8_0 +blk.6.attn_qkv.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 +blk.6.ssm_alpha.weight q8_0 +blk.6.ssm_beta.weight q8_0 +blk.6.ssm_out.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.7.ffn_down.weight q8_0 +blk.7.ffn_gate.weight q8_0 +blk.7.ffn_up.weight q8_0 +blk.8.attn_gate.weight q8_0 +blk.8.attn_qkv.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 +blk.8.ssm_alpha.weight q8_0 +blk.8.ssm_beta.weight q8_0 +blk.8.ssm_out.weight q8_0 +blk.9.attn_gate.weight q8_0 +blk.9.attn_qkv.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 +blk.9.ssm_alpha.weight q8_0 +blk.9.ssm_beta.weight q8_0 +blk.9.ssm_out.weight q8_0 +blk.10.attn_gate.weight q8_0 +blk.10.attn_qkv.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 +blk.10.ssm_alpha.weight q8_0 +blk.10.ssm_beta.weight q8_0 +blk.10.ssm_out.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 +blk.12.attn_gate.weight q8_0 +blk.12.attn_qkv.weight q8_0 +blk.12.ffn_down.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 +blk.12.ssm_alpha.weight q8_0 +blk.12.ssm_beta.weight q8_0 +blk.12.ssm_out.weight q8_0 +blk.13.attn_gate.weight q8_0 +blk.13.attn_qkv.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 +blk.13.ssm_alpha.weight q8_0 +blk.13.ssm_beta.weight q8_0 +blk.13.ssm_out.weight q8_0 +blk.14.attn_gate.weight q8_0 +blk.14.attn_qkv.weight q8_0 +blk.14.ffn_down.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 +blk.14.ssm_alpha.weight q8_0 +blk.14.ssm_beta.weight q8_0 +blk.14.ssm_out.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.15.ffn_down.weight q8_0 +blk.15.ffn_gate.weight q8_0 +blk.15.ffn_up.weight q8_0 +blk.16.attn_gate.weight q8_0 +blk.16.attn_qkv.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 +blk.16.ssm_alpha.weight q8_0 +blk.16.ssm_beta.weight q8_0 +blk.16.ssm_out.weight q8_0 +blk.17.attn_gate.weight q8_0 +blk.17.attn_qkv.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 +blk.17.ssm_alpha.weight q8_0 +blk.17.ssm_beta.weight q8_0 +blk.17.ssm_out.weight q8_0 +blk.18.attn_gate.weight q8_0 +blk.18.attn_qkv.weight q8_0 +blk.18.ffn_down.weight q8_0 +blk.18.ffn_gate.weight q8_0 +blk.18.ffn_up.weight q8_0 +blk.18.ssm_alpha.weight q8_0 +blk.18.ssm_beta.weight q8_0 +blk.18.ssm_out.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 +blk.20.attn_gate.weight q8_0 +blk.20.attn_qkv.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 +blk.20.ssm_alpha.weight q8_0 +blk.20.ssm_beta.weight q8_0 +blk.20.ssm_out.weight q8_0 +blk.21.attn_gate.weight q8_0 +blk.21.attn_qkv.weight q8_0 +blk.21.ffn_down.weight q8_0 +blk.21.ffn_gate.weight q8_0 +blk.21.ffn_up.weight q8_0 +blk.21.ssm_alpha.weight q8_0 +blk.21.ssm_beta.weight q8_0 +blk.21.ssm_out.weight q8_0 +blk.22.attn_gate.weight q8_0 +blk.22.attn_qkv.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 +blk.22.ssm_alpha.weight q8_0 +blk.22.ssm_beta.weight q8_0 +blk.22.ssm_out.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 +blk.24.attn_gate.weight q8_0 +blk.24.attn_qkv.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 +blk.24.ssm_alpha.weight q8_0 +blk.24.ssm_beta.weight q8_0 +blk.24.ssm_out.weight q8_0 +blk.25.attn_gate.weight q8_0 +blk.25.attn_qkv.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 +blk.25.ssm_alpha.weight q8_0 +blk.25.ssm_beta.weight q8_0 +blk.25.ssm_out.weight q8_0 +blk.26.attn_gate.weight q8_0 +blk.26.attn_qkv.weight q8_0 +blk.26.ffn_down.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 +blk.26.ssm_alpha.weight q8_0 +blk.26.ssm_beta.weight q8_0 +blk.26.ssm_out.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 +blk.28.attn_gate.weight q8_0 +blk.28.attn_qkv.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 +blk.28.ssm_alpha.weight q8_0 +blk.28.ssm_beta.weight q8_0 +blk.28.ssm_out.weight q8_0 +blk.29.attn_gate.weight q8_0 +blk.29.attn_qkv.weight q8_0 +blk.29.ffn_down.weight q8_0 +blk.29.ffn_gate.weight q8_0 +blk.29.ffn_up.weight q8_0 +blk.29.ssm_alpha.weight q8_0 +blk.29.ssm_beta.weight q8_0 +blk.29.ssm_out.weight q8_0 +blk.30.attn_gate.weight q8_0 +blk.30.attn_qkv.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 +blk.30.ssm_alpha.weight q8_0 +blk.30.ssm_beta.weight q8_0 +blk.30.ssm_out.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 +blk.32.attn_gate.weight q8_0 +blk.32.attn_qkv.weight q8_0 +blk.32.ffn_down.weight q8_0 +blk.32.ffn_gate.weight q8_0 +blk.32.ffn_up.weight q8_0 +blk.32.ssm_alpha.weight q8_0 +blk.32.ssm_beta.weight q8_0 +blk.32.ssm_out.weight q8_0 +blk.33.attn_gate.weight q8_0 +blk.33.attn_qkv.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 +blk.33.ssm_alpha.weight q8_0 +blk.33.ssm_beta.weight q8_0 +blk.33.ssm_out.weight q8_0 +blk.34.attn_gate.weight q8_0 +blk.34.attn_qkv.weight q8_0 +blk.34.ffn_down.weight q8_0 +blk.34.ffn_gate.weight q8_0 +blk.34.ffn_up.weight q8_0 +blk.34.ssm_alpha.weight q8_0 +blk.34.ssm_beta.weight q8_0 +blk.34.ssm_out.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.35.ffn_down.weight q8_0 +blk.35.ffn_gate.weight q8_0 +blk.35.ffn_up.weight q8_0 +blk.36.attn_gate.weight q8_0 +blk.36.attn_qkv.weight q8_0 +blk.36.ffn_down.weight q8_0 +blk.36.ffn_gate.weight q8_0 +blk.36.ffn_up.weight q8_0 +blk.36.ssm_alpha.weight q8_0 +blk.36.ssm_beta.weight q8_0 +blk.36.ssm_out.weight q8_0 +blk.37.attn_gate.weight q8_0 +blk.37.attn_qkv.weight q8_0 +blk.37.ffn_down.weight q8_0 +blk.37.ffn_gate.weight q8_0 +blk.37.ffn_up.weight q8_0 +blk.37.ssm_alpha.weight q8_0 +blk.37.ssm_beta.weight q8_0 +blk.37.ssm_out.weight q8_0 +blk.38.attn_gate.weight q8_0 +blk.38.attn_qkv.weight q8_0 +blk.38.ffn_down.weight q8_0 +blk.38.ffn_gate.weight q8_0 +blk.38.ffn_up.weight q8_0 +blk.38.ssm_alpha.weight q8_0 +blk.38.ssm_beta.weight q8_0 +blk.38.ssm_out.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.39.ffn_down.weight q8_0 +blk.39.ffn_gate.weight q8_0 +blk.39.ffn_up.weight q8_0 +blk.40.attn_gate.weight q8_0 +blk.40.attn_qkv.weight q8_0 +blk.40.ffn_down.weight q8_0 +blk.40.ffn_gate.weight q8_0 +blk.40.ffn_up.weight q8_0 +blk.40.ssm_alpha.weight q8_0 +blk.40.ssm_beta.weight q8_0 +blk.40.ssm_out.weight q8_0 +blk.41.attn_gate.weight q8_0 +blk.41.attn_qkv.weight q8_0 +blk.41.ffn_down.weight q8_0 +blk.41.ffn_gate.weight q8_0 +blk.41.ffn_up.weight q8_0 +blk.41.ssm_alpha.weight q8_0 +blk.41.ssm_beta.weight q8_0 +blk.41.ssm_out.weight q8_0 +blk.42.attn_gate.weight q8_0 +blk.42.attn_qkv.weight q8_0 +blk.42.ffn_down.weight q8_0 +blk.42.ffn_gate.weight q8_0 +blk.42.ffn_up.weight q8_0 +blk.42.ssm_alpha.weight q8_0 +blk.42.ssm_beta.weight q8_0 +blk.42.ssm_out.weight q8_0 +blk.43.attn_k.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q.weight q8_0 +blk.43.attn_v.weight q8_0 +blk.43.ffn_down.weight q8_0 +blk.43.ffn_gate.weight q8_0 +blk.43.ffn_up.weight q8_0 +blk.44.attn_gate.weight q8_0 +blk.44.attn_qkv.weight q8_0 +blk.44.ffn_down.weight q8_0 +blk.44.ffn_gate.weight q8_0 +blk.44.ffn_up.weight q8_0 +blk.44.ssm_alpha.weight q8_0 +blk.44.ssm_beta.weight q8_0 +blk.44.ssm_out.weight q8_0 +blk.45.attn_gate.weight q8_0 +blk.45.attn_qkv.weight q8_0 +blk.45.ffn_down.weight q8_0 +blk.45.ffn_gate.weight q8_0 +blk.45.ffn_up.weight q8_0 +blk.45.ssm_alpha.weight q8_0 +blk.45.ssm_beta.weight q8_0 +blk.45.ssm_out.weight q8_0 +blk.46.attn_gate.weight q8_0 +blk.46.attn_qkv.weight q8_0 +blk.46.ffn_down.weight q8_0 +blk.46.ffn_gate.weight q8_0 +blk.46.ffn_up.weight q8_0 +blk.46.ssm_alpha.weight q8_0 +blk.46.ssm_beta.weight q8_0 +blk.46.ssm_out.weight q8_0 +blk.47.attn_k.weight q8_0 +blk.47.attn_output.weight q8_0 +blk.47.attn_q.weight q8_0 +blk.47.attn_v.weight q8_0 +blk.47.ffn_down.weight q8_0 +blk.47.ffn_gate.weight q8_0 +blk.47.ffn_up.weight q8_0 +blk.48.attn_gate.weight q8_0 +blk.48.attn_qkv.weight q8_0 +blk.48.ffn_down.weight q8_0 +blk.48.ffn_gate.weight q8_0 +blk.48.ffn_up.weight q8_0 +blk.48.ssm_alpha.weight q8_0 +blk.48.ssm_beta.weight q8_0 +blk.48.ssm_out.weight q8_0 +blk.49.attn_gate.weight q8_0 +blk.49.attn_qkv.weight q8_0 +blk.49.ffn_down.weight q8_0 +blk.49.ffn_gate.weight q8_0 +blk.49.ffn_up.weight q8_0 +blk.49.ssm_alpha.weight q8_0 +blk.49.ssm_beta.weight q8_0 +blk.49.ssm_out.weight q8_0 +blk.50.attn_gate.weight q8_0 +blk.50.attn_qkv.weight q8_0 +blk.50.ffn_down.weight q8_0 +blk.50.ffn_gate.weight q8_0 +blk.50.ffn_up.weight q8_0 +blk.50.ssm_alpha.weight q8_0 +blk.50.ssm_beta.weight q8_0 +blk.50.ssm_out.weight q8_0 +blk.51.attn_k.weight q8_0 +blk.51.attn_output.weight q8_0 +blk.51.attn_q.weight q8_0 +blk.51.attn_v.weight q8_0 +blk.51.ffn_down.weight q8_0 +blk.51.ffn_gate.weight q8_0 +blk.51.ffn_up.weight q8_0 +blk.52.attn_gate.weight q8_0 +blk.52.attn_qkv.weight q8_0 +blk.52.ffn_down.weight q8_0 +blk.52.ffn_gate.weight q8_0 +blk.52.ffn_up.weight q8_0 +blk.52.ssm_alpha.weight q8_0 +blk.52.ssm_beta.weight q8_0 +blk.52.ssm_out.weight q8_0 +blk.53.attn_gate.weight q8_0 +blk.53.attn_qkv.weight q8_0 +blk.53.ffn_down.weight q8_0 +blk.53.ffn_gate.weight q8_0 +blk.53.ffn_up.weight q8_0 +blk.53.ssm_alpha.weight q8_0 +blk.53.ssm_beta.weight q8_0 +blk.53.ssm_out.weight q8_0 +blk.54.attn_gate.weight q8_0 +blk.54.attn_qkv.weight q8_0 +blk.54.ffn_down.weight q8_0 +blk.54.ffn_gate.weight q8_0 +blk.54.ffn_up.weight q8_0 +blk.54.ssm_alpha.weight q8_0 +blk.54.ssm_beta.weight q8_0 +blk.54.ssm_out.weight q8_0 +blk.55.attn_k.weight q8_0 +blk.55.attn_output.weight q8_0 +blk.55.attn_q.weight q8_0 +blk.55.attn_v.weight q8_0 +blk.55.ffn_down.weight q8_0 +blk.55.ffn_gate.weight q8_0 +blk.55.ffn_up.weight q8_0 +blk.56.attn_gate.weight q8_0 +blk.56.attn_qkv.weight q8_0 +blk.56.ffn_down.weight q8_0 +blk.56.ffn_gate.weight q8_0 +blk.56.ffn_up.weight q8_0 +blk.56.ssm_alpha.weight q8_0 +blk.56.ssm_beta.weight q8_0 +blk.56.ssm_out.weight q8_0 +blk.57.attn_gate.weight q8_0 +blk.57.attn_qkv.weight q8_0 +blk.57.ffn_down.weight q8_0 +blk.57.ffn_gate.weight q8_0 +blk.57.ffn_up.weight q8_0 +blk.57.ssm_alpha.weight q8_0 +blk.57.ssm_beta.weight q8_0 +blk.57.ssm_out.weight q8_0 +blk.58.attn_gate.weight q8_0 +blk.58.attn_qkv.weight q8_0 +blk.58.ffn_down.weight q8_0 +blk.58.ffn_gate.weight q8_0 +blk.58.ffn_up.weight q8_0 +blk.58.ssm_alpha.weight q8_0 +blk.58.ssm_beta.weight q8_0 +blk.58.ssm_out.weight q8_0 +blk.59.attn_k.weight q8_0 +blk.59.attn_output.weight q8_0 +blk.59.attn_q.weight q8_0 +blk.59.attn_v.weight q8_0 +blk.59.ffn_down.weight q8_0 +blk.59.ffn_gate.weight q8_0 +blk.59.ffn_up.weight q8_0 +blk.60.attn_gate.weight q8_0 +blk.60.attn_qkv.weight q8_0 +blk.60.ffn_down.weight q8_0 +blk.60.ffn_gate.weight q8_0 +blk.60.ffn_up.weight q8_0 +blk.60.ssm_alpha.weight q8_0 +blk.60.ssm_beta.weight q8_0 +blk.60.ssm_out.weight q8_0 +blk.61.attn_gate.weight q8_0 +blk.61.attn_qkv.weight q8_0 +blk.61.ffn_down.weight q8_0 +blk.61.ffn_gate.weight q8_0 +blk.61.ffn_up.weight q8_0 +blk.61.ssm_alpha.weight q8_0 +blk.61.ssm_beta.weight q8_0 +blk.61.ssm_out.weight q8_0 +blk.62.attn_gate.weight q8_0 +blk.62.attn_qkv.weight q8_0 +blk.62.ffn_down.weight q8_0 +blk.62.ffn_gate.weight q8_0 +blk.62.ffn_up.weight q8_0 +blk.62.ssm_alpha.weight q8_0 +blk.62.ssm_beta.weight q8_0 +blk.62.ssm_out.weight q8_0 +blk.63.attn_k.weight q8_0 +blk.63.attn_output.weight q8_0 +blk.63.attn_q.weight q8_0 +blk.63.attn_v.weight q8_0 +blk.63.ffn_down.weight q8_0 +blk.63.ffn_gate.weight q8_0 +blk.63.ffn_up.weight q8_0 diff --git a/tests/snapshots/qwen3.5-397b-a17b.schema b/tests/snapshots/qwen3.5-397b-a17b.schema new file mode 100644 index 0000000000..801c9b4f16 --- /dev/null +++ b/tests/snapshots/qwen3.5-397b-a17b.schema @@ -0,0 +1,2148 @@ +# Model: Qwen3.5-397B-A17B +# n_embd=4096, n_ff=0, n_vocab=248320, n_layer=60, n_head=32, n_head_kv=2, n_expert=512 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q6_K +blk.0.ffn_down_exps.weight q3_K +blk.0.ffn_down_shexp.weight q3_K +blk.1.ffn_down_exps.weight q3_K +blk.1.ffn_down_shexp.weight q3_K +blk.2.ffn_down_exps.weight q3_K +blk.2.ffn_down_shexp.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K +blk.4.ffn_down_exps.weight q3_K +blk.4.ffn_down_shexp.weight q3_K +blk.5.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_output.weight q3_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.attn_output.weight q3_K +blk.47.attn_v.weight q4_K +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.48.ffn_down_exps.weight q3_K +blk.48.ffn_down_shexp.weight q3_K +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.50.ffn_down_exps.weight q3_K +blk.50.ffn_down_shexp.weight q3_K +blk.51.attn_output.weight q3_K +blk.51.attn_v.weight q4_K +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K +blk.52.ffn_down_exps.weight q3_K +blk.52.ffn_down_shexp.weight q3_K +blk.53.ffn_down_exps.weight q3_K +blk.53.ffn_down_shexp.weight q3_K +blk.54.ffn_down_exps.weight q3_K +blk.54.ffn_down_shexp.weight q3_K +blk.55.attn_output.weight q3_K +blk.55.attn_v.weight q4_K +blk.55.ffn_down_exps.weight q3_K +blk.55.ffn_down_shexp.weight q3_K +blk.56.ffn_down_exps.weight q3_K +blk.56.ffn_down_shexp.weight q3_K +blk.57.ffn_down_exps.weight q3_K +blk.57.ffn_down_shexp.weight q3_K +blk.58.ffn_down_exps.weight q3_K +blk.58.ffn_down_shexp.weight q3_K +blk.59.attn_output.weight q3_K +blk.59.attn_v.weight q4_K +blk.59.ffn_down_exps.weight q3_K +blk.59.ffn_down_shexp.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q4_K +blk.7.ffn_down_shexp.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.8.ffn_down_shexp.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_exps.weight q4_K +blk.9.ffn_down_shexp.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.10.ffn_down_shexp.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q4_K +blk.11.ffn_down_shexp.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down_exps.weight q4_K +blk.12.ffn_down_shexp.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down_exps.weight q4_K +blk.13.ffn_down_shexp.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down_exps.weight q4_K +blk.14.ffn_down_shexp.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down_exps.weight q4_K +blk.16.ffn_down_shexp.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.17.ffn_down_shexp.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down_exps.weight q4_K +blk.18.ffn_down_shexp.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q4_K +blk.19.ffn_down_shexp.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.20.ffn_down_shexp.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down_exps.weight q4_K +blk.21.ffn_down_shexp.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.22.ffn_down_shexp.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.23.ffn_down_shexp.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.24.ffn_down_shexp.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down_exps.weight q4_K +blk.25.ffn_down_shexp.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down_exps.weight q4_K +blk.26.ffn_down_shexp.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down_exps.weight q4_K +blk.28.ffn_down_shexp.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.29.ffn_down_shexp.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down_exps.weight q4_K +blk.30.ffn_down_shexp.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down_exps.weight q4_K +blk.32.ffn_down_shexp.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down_exps.weight q4_K +blk.33.ffn_down_shexp.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.34.ffn_down_shexp.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q4_K +blk.35.ffn_down_shexp.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down_exps.weight q4_K +blk.36.ffn_down_shexp.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down_exps.weight q4_K +blk.37.ffn_down_shexp.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down_exps.weight q4_K +blk.38.ffn_down_shexp.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q4_K +blk.39.ffn_down_shexp.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down_exps.weight q4_K +blk.40.ffn_down_shexp.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down_exps.weight q4_K +blk.41.ffn_down_shexp.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down_exps.weight q4_K +blk.42.ffn_down_shexp.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down_exps.weight q4_K +blk.44.ffn_down_shexp.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down_exps.weight q4_K +blk.45.ffn_down_shexp.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down_exps.weight q4_K +blk.46.ffn_down_shexp.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +blk.47.ffn_down_exps.weight q4_K +blk.47.ffn_down_shexp.weight q4_K +blk.48.attn_qkv.weight q4_K +blk.48.ffn_down_exps.weight q4_K +blk.48.ffn_down_shexp.weight q4_K +blk.49.attn_qkv.weight q4_K +blk.49.ffn_down_exps.weight q4_K +blk.49.ffn_down_shexp.weight q4_K +blk.50.attn_qkv.weight q4_K +blk.50.ffn_down_exps.weight q4_K +blk.50.ffn_down_shexp.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q4_K +blk.51.ffn_down_exps.weight q4_K +blk.51.ffn_down_shexp.weight q4_K +blk.52.attn_qkv.weight q4_K +blk.52.ffn_down_exps.weight q4_K +blk.52.ffn_down_shexp.weight q4_K +blk.53.attn_qkv.weight q4_K +blk.53.ffn_down_exps.weight q4_K +blk.53.ffn_down_shexp.weight q4_K +blk.54.attn_qkv.weight q4_K +blk.54.ffn_down_exps.weight q4_K +blk.54.ffn_down_shexp.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q4_K +blk.55.ffn_down_exps.weight q4_K +blk.55.ffn_down_shexp.weight q4_K +blk.56.attn_qkv.weight q4_K +blk.56.ffn_down_exps.weight q4_K +blk.56.ffn_down_shexp.weight q4_K +blk.57.attn_qkv.weight q4_K +blk.57.ffn_down_exps.weight q4_K +blk.57.ffn_down_shexp.weight q4_K +blk.58.attn_qkv.weight q4_K +blk.58.ffn_down_exps.weight q4_K +blk.58.ffn_down_shexp.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q4_K +blk.59.ffn_down_exps.weight q4_K +blk.59.ffn_down_shexp.weight q4_K + +[Q3_K_L] q3_K +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q5_K +blk.7.ffn_down_shexp.weight q5_K +blk.8.attn_qkv.weight q4_K +blk.8.ffn_down_exps.weight q5_K +blk.8.ffn_down_shexp.weight q5_K +blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_exps.weight q5_K +blk.9.ffn_down_shexp.weight q5_K +blk.10.attn_qkv.weight q4_K +blk.10.ffn_down_exps.weight q5_K +blk.10.ffn_down_shexp.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight q5_K +blk.11.ffn_down_shexp.weight q5_K +blk.12.attn_qkv.weight q4_K +blk.12.ffn_down_exps.weight q5_K +blk.12.ffn_down_shexp.weight q5_K +blk.13.attn_qkv.weight q4_K +blk.13.ffn_down_exps.weight q5_K +blk.13.ffn_down_shexp.weight q5_K +blk.14.attn_qkv.weight q4_K +blk.14.ffn_down_exps.weight q5_K +blk.14.ffn_down_shexp.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K +blk.16.attn_qkv.weight q4_K +blk.16.ffn_down_exps.weight q5_K +blk.16.ffn_down_shexp.weight q5_K +blk.17.attn_qkv.weight q4_K +blk.17.ffn_down_exps.weight q5_K +blk.17.ffn_down_shexp.weight q5_K +blk.18.attn_qkv.weight q4_K +blk.18.ffn_down_exps.weight q5_K +blk.18.ffn_down_shexp.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight q5_K +blk.19.ffn_down_shexp.weight q5_K +blk.20.attn_qkv.weight q4_K +blk.20.ffn_down_exps.weight q5_K +blk.20.ffn_down_shexp.weight q5_K +blk.21.attn_qkv.weight q4_K +blk.21.ffn_down_exps.weight q5_K +blk.21.ffn_down_shexp.weight q5_K +blk.22.attn_qkv.weight q4_K +blk.22.ffn_down_exps.weight q5_K +blk.22.ffn_down_shexp.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.23.ffn_down_shexp.weight q5_K +blk.24.attn_qkv.weight q4_K +blk.24.ffn_down_exps.weight q5_K +blk.24.ffn_down_shexp.weight q5_K +blk.25.attn_qkv.weight q4_K +blk.25.ffn_down_exps.weight q5_K +blk.25.ffn_down_shexp.weight q5_K +blk.26.attn_qkv.weight q4_K +blk.26.ffn_down_exps.weight q5_K +blk.26.ffn_down_shexp.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K +blk.28.attn_qkv.weight q4_K +blk.28.ffn_down_exps.weight q5_K +blk.28.ffn_down_shexp.weight q5_K +blk.29.attn_qkv.weight q4_K +blk.29.ffn_down_exps.weight q5_K +blk.29.ffn_down_shexp.weight q5_K +blk.30.attn_qkv.weight q4_K +blk.30.ffn_down_exps.weight q5_K +blk.30.ffn_down_shexp.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K +blk.32.attn_qkv.weight q4_K +blk.32.ffn_down_exps.weight q5_K +blk.32.ffn_down_shexp.weight q5_K +blk.33.attn_qkv.weight q4_K +blk.33.ffn_down_exps.weight q5_K +blk.33.ffn_down_shexp.weight q5_K +blk.34.attn_qkv.weight q4_K +blk.34.ffn_down_exps.weight q5_K +blk.34.ffn_down_shexp.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight q5_K +blk.35.ffn_down_shexp.weight q5_K +blk.36.attn_qkv.weight q4_K +blk.36.ffn_down_exps.weight q5_K +blk.36.ffn_down_shexp.weight q5_K +blk.37.attn_qkv.weight q4_K +blk.37.ffn_down_exps.weight q5_K +blk.37.ffn_down_shexp.weight q5_K +blk.38.attn_qkv.weight q4_K +blk.38.ffn_down_exps.weight q5_K +blk.38.ffn_down_shexp.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down_exps.weight q5_K +blk.39.ffn_down_shexp.weight q5_K +blk.40.attn_qkv.weight q4_K +blk.40.ffn_down_exps.weight q5_K +blk.40.ffn_down_shexp.weight q5_K +blk.41.attn_qkv.weight q4_K +blk.41.ffn_down_exps.weight q5_K +blk.41.ffn_down_shexp.weight q5_K +blk.42.attn_qkv.weight q4_K +blk.42.ffn_down_exps.weight q5_K +blk.42.ffn_down_shexp.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K +blk.44.attn_qkv.weight q4_K +blk.44.ffn_down_exps.weight q5_K +blk.44.ffn_down_shexp.weight q5_K +blk.45.attn_qkv.weight q4_K +blk.45.ffn_down_exps.weight q5_K +blk.45.ffn_down_shexp.weight q5_K +blk.46.attn_qkv.weight q4_K +blk.46.ffn_down_exps.weight q5_K +blk.46.ffn_down_shexp.weight q5_K +blk.47.attn_output.weight q5_K +blk.47.attn_v.weight q5_K +blk.47.ffn_down_exps.weight q5_K +blk.47.ffn_down_shexp.weight q5_K +blk.48.attn_qkv.weight q4_K +blk.48.ffn_down_exps.weight q5_K +blk.48.ffn_down_shexp.weight q5_K +blk.49.attn_qkv.weight q4_K +blk.49.ffn_down_exps.weight q5_K +blk.49.ffn_down_shexp.weight q5_K +blk.50.attn_qkv.weight q4_K +blk.50.ffn_down_exps.weight q5_K +blk.50.ffn_down_shexp.weight q5_K +blk.51.attn_output.weight q5_K +blk.51.attn_v.weight q5_K +blk.51.ffn_down_exps.weight q5_K +blk.51.ffn_down_shexp.weight q5_K +blk.52.attn_qkv.weight q4_K +blk.52.ffn_down_exps.weight q5_K +blk.52.ffn_down_shexp.weight q5_K +blk.53.attn_qkv.weight q4_K +blk.53.ffn_down_exps.weight q5_K +blk.53.ffn_down_shexp.weight q5_K +blk.54.attn_qkv.weight q4_K +blk.54.ffn_down_exps.weight q5_K +blk.54.ffn_down_shexp.weight q5_K +blk.55.attn_output.weight q5_K +blk.55.attn_v.weight q5_K +blk.55.ffn_down_exps.weight q5_K +blk.55.ffn_down_shexp.weight q5_K +blk.56.attn_qkv.weight q4_K +blk.56.ffn_down_exps.weight q5_K +blk.56.ffn_down_shexp.weight q5_K +blk.57.attn_qkv.weight q4_K +blk.57.ffn_down_exps.weight q5_K +blk.57.ffn_down_shexp.weight q5_K +blk.58.attn_qkv.weight q4_K +blk.58.ffn_down_exps.weight q5_K +blk.58.ffn_down_shexp.weight q5_K +blk.59.attn_output.weight q5_K +blk.59.attn_v.weight q5_K +blk.59.ffn_down_exps.weight q5_K +blk.59.ffn_down_shexp.weight q5_K + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K + +[Q4_K_M] q4_K +output.weight q6_K +blk.0.attn_qkv.weight q5_K +blk.0.ffn_down_exps.weight q6_K +blk.0.ffn_down_shexp.weight q6_K +blk.1.attn_qkv.weight q5_K +blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K +blk.2.attn_qkv.weight q5_K +blk.2.ffn_down_exps.weight q6_K +blk.2.ffn_down_shexp.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.attn_qkv.weight q5_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.5.attn_qkv.weight q5_K +blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K +blk.6.attn_qkv.weight q5_K +blk.6.ffn_down_exps.weight q6_K +blk.6.ffn_down_shexp.weight q6_K +blk.7.attn_v.weight q6_K +blk.8.attn_qkv.weight q5_K +blk.9.attn_qkv.weight q5_K +blk.9.ffn_down_exps.weight q6_K +blk.9.ffn_down_shexp.weight q6_K +blk.10.attn_qkv.weight q5_K +blk.11.attn_v.weight q6_K +blk.12.attn_qkv.weight q5_K +blk.12.ffn_down_exps.weight q6_K +blk.12.ffn_down_shexp.weight q6_K +blk.13.attn_qkv.weight q5_K +blk.14.attn_qkv.weight q5_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.15.ffn_down_shexp.weight q6_K +blk.16.attn_qkv.weight q5_K +blk.17.attn_qkv.weight q5_K +blk.18.attn_qkv.weight q5_K +blk.18.ffn_down_exps.weight q6_K +blk.18.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.20.attn_qkv.weight q5_K +blk.21.attn_qkv.weight q5_K +blk.21.ffn_down_exps.weight q6_K +blk.21.ffn_down_shexp.weight q6_K +blk.22.attn_qkv.weight q5_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q5_K +blk.24.ffn_down_exps.weight q6_K +blk.24.ffn_down_shexp.weight q6_K +blk.25.attn_qkv.weight q5_K +blk.26.attn_qkv.weight q5_K +blk.27.attn_v.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.27.ffn_down_shexp.weight q6_K +blk.28.attn_qkv.weight q5_K +blk.29.attn_qkv.weight q5_K +blk.30.attn_qkv.weight q5_K +blk.30.ffn_down_exps.weight q6_K +blk.30.ffn_down_shexp.weight q6_K +blk.32.attn_qkv.weight q5_K +blk.33.attn_qkv.weight q5_K +blk.33.ffn_down_exps.weight q6_K +blk.33.ffn_down_shexp.weight q6_K +blk.34.attn_qkv.weight q5_K +blk.36.attn_qkv.weight q5_K +blk.36.ffn_down_exps.weight q6_K +blk.36.ffn_down_shexp.weight q6_K +blk.37.attn_qkv.weight q5_K +blk.38.attn_qkv.weight q5_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.40.attn_qkv.weight q5_K +blk.41.attn_qkv.weight q5_K +blk.42.attn_qkv.weight q5_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.44.attn_qkv.weight q5_K +blk.45.attn_qkv.weight q5_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.46.attn_qkv.weight q5_K +blk.48.attn_qkv.weight q5_K +blk.48.ffn_down_exps.weight q6_K +blk.48.ffn_down_shexp.weight q6_K +blk.49.attn_qkv.weight q5_K +blk.50.attn_qkv.weight q5_K +blk.51.attn_v.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K +blk.52.attn_qkv.weight q5_K +blk.52.ffn_down_exps.weight q6_K +blk.52.ffn_down_shexp.weight q6_K +blk.53.attn_qkv.weight q5_K +blk.53.ffn_down_exps.weight q6_K +blk.53.ffn_down_shexp.weight q6_K +blk.54.attn_qkv.weight q5_K +blk.54.ffn_down_exps.weight q6_K +blk.54.ffn_down_shexp.weight q6_K +blk.55.ffn_down_exps.weight q6_K +blk.55.ffn_down_shexp.weight q6_K +blk.56.attn_qkv.weight q5_K +blk.56.ffn_down_exps.weight q6_K +blk.56.ffn_down_shexp.weight q6_K +blk.57.attn_qkv.weight q5_K +blk.57.ffn_down_exps.weight q6_K +blk.57.ffn_down_shexp.weight q6_K +blk.58.attn_qkv.weight q5_K +blk.58.ffn_down_exps.weight q6_K +blk.58.ffn_down_shexp.weight q6_K +blk.59.ffn_down_exps.weight q6_K +blk.59.ffn_down_shexp.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +output.weight q6_K +blk.0.attn_qkv.weight q6_K +blk.0.ffn_down_exps.weight q6_K +blk.0.ffn_down_shexp.weight q6_K +blk.1.attn_qkv.weight q6_K +blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K +blk.2.attn_qkv.weight q6_K +blk.2.ffn_down_exps.weight q6_K +blk.2.ffn_down_shexp.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.attn_qkv.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.5.attn_qkv.weight q6_K +blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K +blk.6.attn_qkv.weight q6_K +blk.6.ffn_down_exps.weight q6_K +blk.6.ffn_down_shexp.weight q6_K +blk.7.attn_v.weight q6_K +blk.8.attn_qkv.weight q6_K +blk.9.attn_qkv.weight q6_K +blk.9.ffn_down_exps.weight q6_K +blk.9.ffn_down_shexp.weight q6_K +blk.10.attn_qkv.weight q6_K +blk.11.attn_v.weight q6_K +blk.12.attn_qkv.weight q6_K +blk.12.ffn_down_exps.weight q6_K +blk.12.ffn_down_shexp.weight q6_K +blk.13.attn_qkv.weight q6_K +blk.14.attn_qkv.weight q6_K +blk.15.attn_v.weight q6_K +blk.15.ffn_down_exps.weight q6_K +blk.15.ffn_down_shexp.weight q6_K +blk.16.attn_qkv.weight q6_K +blk.17.attn_qkv.weight q6_K +blk.18.attn_qkv.weight q6_K +blk.18.ffn_down_exps.weight q6_K +blk.18.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.20.attn_qkv.weight q6_K +blk.21.attn_qkv.weight q6_K +blk.21.ffn_down_exps.weight q6_K +blk.21.ffn_down_shexp.weight q6_K +blk.22.attn_qkv.weight q6_K +blk.23.attn_v.weight q6_K +blk.24.attn_qkv.weight q6_K +blk.24.ffn_down_exps.weight q6_K +blk.24.ffn_down_shexp.weight q6_K +blk.25.attn_qkv.weight q6_K +blk.26.attn_qkv.weight q6_K +blk.27.attn_v.weight q6_K +blk.27.ffn_down_exps.weight q6_K +blk.27.ffn_down_shexp.weight q6_K +blk.28.attn_qkv.weight q6_K +blk.29.attn_qkv.weight q6_K +blk.30.attn_qkv.weight q6_K +blk.30.ffn_down_exps.weight q6_K +blk.30.ffn_down_shexp.weight q6_K +blk.32.attn_qkv.weight q6_K +blk.33.attn_qkv.weight q6_K +blk.33.ffn_down_exps.weight q6_K +blk.33.ffn_down_shexp.weight q6_K +blk.34.attn_qkv.weight q6_K +blk.36.attn_qkv.weight q6_K +blk.36.ffn_down_exps.weight q6_K +blk.36.ffn_down_shexp.weight q6_K +blk.37.attn_qkv.weight q6_K +blk.38.attn_qkv.weight q6_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.40.attn_qkv.weight q6_K +blk.41.attn_qkv.weight q6_K +blk.42.attn_qkv.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.44.attn_qkv.weight q6_K +blk.45.attn_qkv.weight q6_K +blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K +blk.46.attn_qkv.weight q6_K +blk.48.attn_qkv.weight q6_K +blk.48.ffn_down_exps.weight q6_K +blk.48.ffn_down_shexp.weight q6_K +blk.49.attn_qkv.weight q6_K +blk.50.attn_qkv.weight q6_K +blk.51.attn_v.weight q6_K +blk.51.ffn_down_exps.weight q6_K +blk.51.ffn_down_shexp.weight q6_K +blk.52.attn_qkv.weight q6_K +blk.52.ffn_down_exps.weight q6_K +blk.52.ffn_down_shexp.weight q6_K +blk.53.attn_qkv.weight q6_K +blk.53.ffn_down_exps.weight q6_K +blk.53.ffn_down_shexp.weight q6_K +blk.54.attn_qkv.weight q6_K +blk.54.ffn_down_exps.weight q6_K +blk.54.ffn_down_shexp.weight q6_K +blk.55.ffn_down_exps.weight q6_K +blk.55.ffn_down_shexp.weight q6_K +blk.56.attn_qkv.weight q6_K +blk.56.ffn_down_exps.weight q6_K +blk.56.ffn_down_shexp.weight q6_K +blk.57.attn_qkv.weight q6_K +blk.57.ffn_down_exps.weight q6_K +blk.57.ffn_down_shexp.weight q6_K +blk.58.attn_qkv.weight q6_K +blk.58.ffn_down_exps.weight q6_K +blk.58.ffn_down_shexp.weight q6_K +blk.59.ffn_down_exps.weight q6_K +blk.59.ffn_down_shexp.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q6_K +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K + +[IQ3_XS] iq3_s +output.weight q6_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.9.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.12.ffn_up_shexp.weight iq3_xxs +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_gate_shexp.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs +blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs +blk.32.ffn_up_exps.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs +blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs +blk.33.ffn_up_exps.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs +blk.34.ffn_gate_exps.weight iq3_xxs +blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K +blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs +blk.35.ffn_up_exps.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs +blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs +blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs +blk.37.ffn_up_exps.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q4_K +blk.39.ffn_gate_exps.weight iq3_xxs +blk.39.ffn_gate_shexp.weight iq3_xxs +blk.39.ffn_up_exps.weight iq3_xxs +blk.39.ffn_up_shexp.weight iq3_xxs +blk.40.ffn_gate_exps.weight iq3_xxs +blk.40.ffn_gate_shexp.weight iq3_xxs +blk.40.ffn_up_exps.weight iq3_xxs +blk.40.ffn_up_shexp.weight iq3_xxs +blk.41.ffn_gate_exps.weight iq3_xxs +blk.41.ffn_gate_shexp.weight iq3_xxs +blk.41.ffn_up_exps.weight iq3_xxs +blk.41.ffn_up_shexp.weight iq3_xxs +blk.42.ffn_gate_exps.weight iq3_xxs +blk.42.ffn_gate_shexp.weight iq3_xxs +blk.42.ffn_up_exps.weight iq3_xxs +blk.42.ffn_up_shexp.weight iq3_xxs +blk.43.attn_k.weight iq3_xxs +blk.43.attn_q.weight iq3_xxs +blk.43.attn_v.weight q4_K +blk.43.ffn_gate_exps.weight iq3_xxs +blk.43.ffn_gate_shexp.weight iq3_xxs +blk.43.ffn_up_exps.weight iq3_xxs +blk.43.ffn_up_shexp.weight iq3_xxs +blk.44.ffn_gate_exps.weight iq3_xxs +blk.44.ffn_gate_shexp.weight iq3_xxs +blk.44.ffn_up_exps.weight iq3_xxs +blk.44.ffn_up_shexp.weight iq3_xxs +blk.45.ffn_gate_exps.weight iq3_xxs +blk.45.ffn_gate_shexp.weight iq3_xxs +blk.45.ffn_up_exps.weight iq3_xxs +blk.45.ffn_up_shexp.weight iq3_xxs +blk.46.ffn_gate_exps.weight iq3_xxs +blk.46.ffn_gate_shexp.weight iq3_xxs +blk.46.ffn_up_exps.weight iq3_xxs +blk.46.ffn_up_shexp.weight iq3_xxs +blk.47.attn_k.weight iq3_xxs +blk.47.attn_q.weight iq3_xxs +blk.47.attn_v.weight q4_K +blk.47.ffn_gate_exps.weight iq3_xxs +blk.47.ffn_gate_shexp.weight iq3_xxs +blk.47.ffn_up_exps.weight iq3_xxs +blk.47.ffn_up_shexp.weight iq3_xxs +blk.48.ffn_gate_exps.weight iq3_xxs +blk.48.ffn_gate_shexp.weight iq3_xxs +blk.48.ffn_up_exps.weight iq3_xxs +blk.48.ffn_up_shexp.weight iq3_xxs +blk.49.ffn_gate_exps.weight iq3_xxs +blk.49.ffn_gate_shexp.weight iq3_xxs +blk.49.ffn_up_exps.weight iq3_xxs +blk.49.ffn_up_shexp.weight iq3_xxs +blk.50.ffn_gate_exps.weight iq3_xxs +blk.50.ffn_gate_shexp.weight iq3_xxs +blk.50.ffn_up_exps.weight iq3_xxs +blk.50.ffn_up_shexp.weight iq3_xxs +blk.51.attn_k.weight iq3_xxs +blk.51.attn_q.weight iq3_xxs +blk.51.attn_v.weight q4_K +blk.51.ffn_gate_exps.weight iq3_xxs +blk.51.ffn_gate_shexp.weight iq3_xxs +blk.51.ffn_up_exps.weight iq3_xxs +blk.51.ffn_up_shexp.weight iq3_xxs +blk.55.attn_k.weight iq3_xxs +blk.55.attn_q.weight iq3_xxs +blk.55.attn_v.weight q4_K +blk.59.attn_k.weight iq3_xxs +blk.59.attn_q.weight iq3_xxs +blk.59.attn_v.weight q4_K + +[IQ3_XXS] iq3_xxs +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_k.weight iq2_s +blk.43.attn_output.weight iq3_s +blk.43.attn_q.weight iq2_s +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K +blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K +blk.46.ffn_down_exps.weight q3_K +blk.46.ffn_down_shexp.weight q3_K +blk.47.attn_k.weight iq2_s +blk.47.attn_output.weight iq3_s +blk.47.attn_q.weight iq2_s +blk.47.attn_v.weight q4_K +blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K +blk.48.ffn_down_exps.weight q3_K +blk.48.ffn_down_shexp.weight q3_K +blk.49.ffn_down_exps.weight q3_K +blk.49.ffn_down_shexp.weight q3_K +blk.50.ffn_down_exps.weight q3_K +blk.50.ffn_down_shexp.weight q3_K +blk.51.attn_k.weight iq2_s +blk.51.attn_output.weight iq3_s +blk.51.attn_q.weight iq2_s +blk.51.attn_v.weight q4_K +blk.51.ffn_down_exps.weight q3_K +blk.51.ffn_down_shexp.weight q3_K +blk.52.ffn_down_exps.weight q3_K +blk.52.ffn_down_shexp.weight q3_K +blk.53.ffn_down_exps.weight q3_K +blk.53.ffn_down_shexp.weight q3_K +blk.54.ffn_down_exps.weight q3_K +blk.54.ffn_down_shexp.weight q3_K +blk.55.attn_k.weight iq2_s +blk.55.attn_output.weight iq3_s +blk.55.attn_q.weight iq2_s +blk.55.attn_v.weight q4_K +blk.55.ffn_down_exps.weight q3_K +blk.55.ffn_down_shexp.weight q3_K +blk.56.ffn_down_exps.weight q3_K +blk.56.ffn_down_shexp.weight q3_K +blk.57.ffn_down_exps.weight q3_K +blk.57.ffn_down_shexp.weight q3_K +blk.58.ffn_down_exps.weight q3_K +blk.58.ffn_down_shexp.weight q3_K +blk.59.attn_k.weight iq2_s +blk.59.attn_output.weight iq3_s +blk.59.attn_q.weight iq2_s +blk.59.attn_v.weight q4_K +blk.59.ffn_down_exps.weight q3_K +blk.59.ffn_down_shexp.weight q3_K + +[IQ1_S] iq1_s +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.59.attn_v.weight q5_K + +[IQ3_S] iq3_s +output.weight q6_K +blk.3.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.47.attn_v.weight q4_K +blk.51.attn_v.weight q4_K +blk.55.attn_v.weight q4_K +blk.59.attn_v.weight q4_K + +[IQ3_M] iq3_s +output.weight q6_K +blk.0.attn_qkv.weight q4_K +blk.0.ffn_down_exps.weight q4_K +blk.0.ffn_down_shexp.weight q4_K +blk.1.attn_qkv.weight q4_K +blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K +blk.2.attn_qkv.weight q4_K +blk.2.ffn_down_exps.weight q4_K +blk.2.ffn_down_shexp.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_qkv.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_qkv.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_qkv.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_qkv.weight q4_K +blk.9.attn_qkv.weight q4_K +blk.10.attn_qkv.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_qkv.weight q4_K +blk.13.attn_qkv.weight q4_K +blk.14.attn_qkv.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_qkv.weight q4_K +blk.17.attn_qkv.weight q4_K +blk.18.attn_qkv.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_qkv.weight q4_K +blk.21.attn_qkv.weight q4_K +blk.22.attn_qkv.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_qkv.weight q4_K +blk.25.attn_qkv.weight q4_K +blk.26.attn_qkv.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_qkv.weight q4_K +blk.29.attn_qkv.weight q4_K +blk.30.attn_qkv.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_qkv.weight q4_K +blk.33.attn_qkv.weight q4_K +blk.34.attn_qkv.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_qkv.weight q4_K +blk.37.attn_qkv.weight q4_K +blk.38.attn_qkv.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_qkv.weight q4_K +blk.41.attn_qkv.weight q4_K +blk.42.attn_qkv.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_qkv.weight q4_K +blk.45.attn_qkv.weight q4_K +blk.46.attn_qkv.weight q4_K +blk.47.attn_output.weight q4_K +blk.47.attn_v.weight q4_K +blk.48.attn_qkv.weight q4_K +blk.49.attn_qkv.weight q4_K +blk.50.attn_qkv.weight q4_K +blk.51.attn_output.weight q4_K +blk.51.attn_v.weight q4_K +blk.52.attn_qkv.weight q4_K +blk.53.attn_qkv.weight q4_K +blk.54.attn_qkv.weight q4_K +blk.55.attn_output.weight q4_K +blk.55.attn_v.weight q4_K +blk.56.attn_qkv.weight q4_K +blk.57.attn_qkv.weight q4_K +blk.58.attn_qkv.weight q4_K +blk.59.attn_output.weight q4_K +blk.59.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight iq3_s +blk.0.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.2.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q5_K +token_embd.weight iq3_s +blk.0.ffn_down_exps.weight iq3_s +blk.0.ffn_down_shexp.weight iq3_s +blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s +blk.2.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq3_s +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq3_s +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq3_s +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq3_s +blk.59.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.ffn_down_exps.weight q5_K +blk.0.ffn_down_shexp.weight q5_K +blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K +blk.2.ffn_down_exps.weight q5_K +blk.2.ffn_down_shexp.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.47.attn_v.weight q5_K +blk.51.attn_v.weight q5_K +blk.55.attn_v.weight q5_K +blk.59.attn_v.weight q5_K + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.ffn_down_exps.weight q2_K +blk.0.ffn_down_shexp.weight q2_K +blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K +blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.47.attn_output.weight iq2_xxs +blk.47.attn_v.weight q4_K +blk.51.attn_output.weight iq2_xxs +blk.51.attn_v.weight q4_K +blk.55.attn_output.weight iq2_xxs +blk.55.attn_v.weight q4_K +blk.59.attn_output.weight iq2_xxs +blk.59.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q6_K +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q6_K +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_gate.weight q8_0 +blk.0.attn_qkv.weight q8_0 +blk.0.ffn_down_shexp.weight q8_0 +blk.0.ffn_gate_shexp.weight q8_0 +blk.0.ffn_up_shexp.weight q8_0 +blk.0.ssm_alpha.weight q8_0 +blk.0.ssm_beta.weight q8_0 +blk.0.ssm_out.weight q8_0 +blk.1.attn_gate.weight q8_0 +blk.1.attn_qkv.weight q8_0 +blk.1.ffn_down_shexp.weight q8_0 +blk.1.ffn_gate_shexp.weight q8_0 +blk.1.ffn_up_shexp.weight q8_0 +blk.1.ssm_alpha.weight q8_0 +blk.1.ssm_beta.weight q8_0 +blk.1.ssm_out.weight q8_0 +blk.2.attn_gate.weight q8_0 +blk.2.attn_qkv.weight q8_0 +blk.2.ffn_down_shexp.weight q8_0 +blk.2.ffn_gate_shexp.weight q8_0 +blk.2.ffn_up_shexp.weight q8_0 +blk.2.ssm_alpha.weight q8_0 +blk.2.ssm_beta.weight q8_0 +blk.2.ssm_out.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 +blk.4.attn_gate.weight q8_0 +blk.4.attn_qkv.weight q8_0 +blk.4.ffn_down_shexp.weight q8_0 +blk.4.ffn_gate_shexp.weight q8_0 +blk.4.ffn_up_shexp.weight q8_0 +blk.4.ssm_alpha.weight q8_0 +blk.4.ssm_beta.weight q8_0 +blk.4.ssm_out.weight q8_0 +blk.5.attn_gate.weight q8_0 +blk.5.attn_qkv.weight q8_0 +blk.5.ffn_down_shexp.weight q8_0 +blk.5.ffn_gate_shexp.weight q8_0 +blk.5.ffn_up_shexp.weight q8_0 +blk.5.ssm_alpha.weight q8_0 +blk.5.ssm_beta.weight q8_0 +blk.5.ssm_out.weight q8_0 +blk.6.attn_gate.weight q8_0 +blk.6.attn_qkv.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_gate_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 +blk.6.ssm_alpha.weight q8_0 +blk.6.ssm_beta.weight q8_0 +blk.6.ssm_out.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 +blk.8.attn_gate.weight q8_0 +blk.8.attn_qkv.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_gate_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 +blk.8.ssm_alpha.weight q8_0 +blk.8.ssm_beta.weight q8_0 +blk.8.ssm_out.weight q8_0 +blk.9.attn_gate.weight q8_0 +blk.9.attn_qkv.weight q8_0 +blk.9.ffn_down_shexp.weight q8_0 +blk.9.ffn_gate_shexp.weight q8_0 +blk.9.ffn_up_shexp.weight q8_0 +blk.9.ssm_alpha.weight q8_0 +blk.9.ssm_beta.weight q8_0 +blk.9.ssm_out.weight q8_0 +blk.10.attn_gate.weight q8_0 +blk.10.attn_qkv.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_gate_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 +blk.10.ssm_alpha.weight q8_0 +blk.10.ssm_beta.weight q8_0 +blk.10.ssm_out.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 +blk.12.attn_gate.weight q8_0 +blk.12.attn_qkv.weight q8_0 +blk.12.ffn_down_shexp.weight q8_0 +blk.12.ffn_gate_shexp.weight q8_0 +blk.12.ffn_up_shexp.weight q8_0 +blk.12.ssm_alpha.weight q8_0 +blk.12.ssm_beta.weight q8_0 +blk.12.ssm_out.weight q8_0 +blk.13.attn_gate.weight q8_0 +blk.13.attn_qkv.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_gate_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 +blk.13.ssm_alpha.weight q8_0 +blk.13.ssm_beta.weight q8_0 +blk.13.ssm_out.weight q8_0 +blk.14.attn_gate.weight q8_0 +blk.14.attn_qkv.weight q8_0 +blk.14.ffn_down_shexp.weight q8_0 +blk.14.ffn_gate_shexp.weight q8_0 +blk.14.ffn_up_shexp.weight q8_0 +blk.14.ssm_alpha.weight q8_0 +blk.14.ssm_beta.weight q8_0 +blk.14.ssm_out.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 +blk.16.attn_gate.weight q8_0 +blk.16.attn_qkv.weight q8_0 +blk.16.ffn_down_shexp.weight q8_0 +blk.16.ffn_gate_shexp.weight q8_0 +blk.16.ffn_up_shexp.weight q8_0 +blk.16.ssm_alpha.weight q8_0 +blk.16.ssm_beta.weight q8_0 +blk.16.ssm_out.weight q8_0 +blk.17.attn_gate.weight q8_0 +blk.17.attn_qkv.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_gate_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 +blk.17.ssm_alpha.weight q8_0 +blk.17.ssm_beta.weight q8_0 +blk.17.ssm_out.weight q8_0 +blk.18.attn_gate.weight q8_0 +blk.18.attn_qkv.weight q8_0 +blk.18.ffn_down_shexp.weight q8_0 +blk.18.ffn_gate_shexp.weight q8_0 +blk.18.ffn_up_shexp.weight q8_0 +blk.18.ssm_alpha.weight q8_0 +blk.18.ssm_beta.weight q8_0 +blk.18.ssm_out.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 +blk.20.attn_gate.weight q8_0 +blk.20.attn_qkv.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_gate_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 +blk.20.ssm_alpha.weight q8_0 +blk.20.ssm_beta.weight q8_0 +blk.20.ssm_out.weight q8_0 +blk.21.attn_gate.weight q8_0 +blk.21.attn_qkv.weight q8_0 +blk.21.ffn_down_shexp.weight q8_0 +blk.21.ffn_gate_shexp.weight q8_0 +blk.21.ffn_up_shexp.weight q8_0 +blk.21.ssm_alpha.weight q8_0 +blk.21.ssm_beta.weight q8_0 +blk.21.ssm_out.weight q8_0 +blk.22.attn_gate.weight q8_0 +blk.22.attn_qkv.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_gate_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 +blk.22.ssm_alpha.weight q8_0 +blk.22.ssm_beta.weight q8_0 +blk.22.ssm_out.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 +blk.24.attn_gate.weight q8_0 +blk.24.attn_qkv.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_gate_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 +blk.24.ssm_alpha.weight q8_0 +blk.24.ssm_beta.weight q8_0 +blk.24.ssm_out.weight q8_0 +blk.25.attn_gate.weight q8_0 +blk.25.attn_qkv.weight q8_0 +blk.25.ffn_down_shexp.weight q8_0 +blk.25.ffn_gate_shexp.weight q8_0 +blk.25.ffn_up_shexp.weight q8_0 +blk.25.ssm_alpha.weight q8_0 +blk.25.ssm_beta.weight q8_0 +blk.25.ssm_out.weight q8_0 +blk.26.attn_gate.weight q8_0 +blk.26.attn_qkv.weight q8_0 +blk.26.ffn_down_shexp.weight q8_0 +blk.26.ffn_gate_shexp.weight q8_0 +blk.26.ffn_up_shexp.weight q8_0 +blk.26.ssm_alpha.weight q8_0 +blk.26.ssm_beta.weight q8_0 +blk.26.ssm_out.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 +blk.28.attn_gate.weight q8_0 +blk.28.attn_qkv.weight q8_0 +blk.28.ffn_down_shexp.weight q8_0 +blk.28.ffn_gate_shexp.weight q8_0 +blk.28.ffn_up_shexp.weight q8_0 +blk.28.ssm_alpha.weight q8_0 +blk.28.ssm_beta.weight q8_0 +blk.28.ssm_out.weight q8_0 +blk.29.attn_gate.weight q8_0 +blk.29.attn_qkv.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_gate_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 +blk.29.ssm_alpha.weight q8_0 +blk.29.ssm_beta.weight q8_0 +blk.29.ssm_out.weight q8_0 +blk.30.attn_gate.weight q8_0 +blk.30.attn_qkv.weight q8_0 +blk.30.ffn_down_shexp.weight q8_0 +blk.30.ffn_gate_shexp.weight q8_0 +blk.30.ffn_up_shexp.weight q8_0 +blk.30.ssm_alpha.weight q8_0 +blk.30.ssm_beta.weight q8_0 +blk.30.ssm_out.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 +blk.32.attn_gate.weight q8_0 +blk.32.attn_qkv.weight q8_0 +blk.32.ffn_down_shexp.weight q8_0 +blk.32.ffn_gate_shexp.weight q8_0 +blk.32.ffn_up_shexp.weight q8_0 +blk.32.ssm_alpha.weight q8_0 +blk.32.ssm_beta.weight q8_0 +blk.32.ssm_out.weight q8_0 +blk.33.attn_gate.weight q8_0 +blk.33.attn_qkv.weight q8_0 +blk.33.ffn_down_shexp.weight q8_0 +blk.33.ffn_gate_shexp.weight q8_0 +blk.33.ffn_up_shexp.weight q8_0 +blk.33.ssm_alpha.weight q8_0 +blk.33.ssm_beta.weight q8_0 +blk.33.ssm_out.weight q8_0 +blk.34.attn_gate.weight q8_0 +blk.34.attn_qkv.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_gate_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 +blk.34.ssm_alpha.weight q8_0 +blk.34.ssm_beta.weight q8_0 +blk.34.ssm_out.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 +blk.36.attn_gate.weight q8_0 +blk.36.attn_qkv.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_gate_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 +blk.36.ssm_alpha.weight q8_0 +blk.36.ssm_beta.weight q8_0 +blk.36.ssm_out.weight q8_0 +blk.37.attn_gate.weight q8_0 +blk.37.attn_qkv.weight q8_0 +blk.37.ffn_down_shexp.weight q8_0 +blk.37.ffn_gate_shexp.weight q8_0 +blk.37.ffn_up_shexp.weight q8_0 +blk.37.ssm_alpha.weight q8_0 +blk.37.ssm_beta.weight q8_0 +blk.37.ssm_out.weight q8_0 +blk.38.attn_gate.weight q8_0 +blk.38.attn_qkv.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_gate_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 +blk.38.ssm_alpha.weight q8_0 +blk.38.ssm_beta.weight q8_0 +blk.38.ssm_out.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 +blk.40.attn_gate.weight q8_0 +blk.40.attn_qkv.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_gate_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 +blk.40.ssm_alpha.weight q8_0 +blk.40.ssm_beta.weight q8_0 +blk.40.ssm_out.weight q8_0 +blk.41.attn_gate.weight q8_0 +blk.41.attn_qkv.weight q8_0 +blk.41.ffn_down_shexp.weight q8_0 +blk.41.ffn_gate_shexp.weight q8_0 +blk.41.ffn_up_shexp.weight q8_0 +blk.41.ssm_alpha.weight q8_0 +blk.41.ssm_beta.weight q8_0 +blk.41.ssm_out.weight q8_0 +blk.42.attn_gate.weight q8_0 +blk.42.attn_qkv.weight q8_0 +blk.42.ffn_down_shexp.weight q8_0 +blk.42.ffn_gate_shexp.weight q8_0 +blk.42.ffn_up_shexp.weight q8_0 +blk.42.ssm_alpha.weight q8_0 +blk.42.ssm_beta.weight q8_0 +blk.42.ssm_out.weight q8_0 +blk.43.attn_k.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q.weight q8_0 +blk.43.attn_v.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 +blk.44.attn_gate.weight q8_0 +blk.44.attn_qkv.weight q8_0 +blk.44.ffn_down_shexp.weight q8_0 +blk.44.ffn_gate_shexp.weight q8_0 +blk.44.ffn_up_shexp.weight q8_0 +blk.44.ssm_alpha.weight q8_0 +blk.44.ssm_beta.weight q8_0 +blk.44.ssm_out.weight q8_0 +blk.45.attn_gate.weight q8_0 +blk.45.attn_qkv.weight q8_0 +blk.45.ffn_down_shexp.weight q8_0 +blk.45.ffn_gate_shexp.weight q8_0 +blk.45.ffn_up_shexp.weight q8_0 +blk.45.ssm_alpha.weight q8_0 +blk.45.ssm_beta.weight q8_0 +blk.45.ssm_out.weight q8_0 +blk.46.attn_gate.weight q8_0 +blk.46.attn_qkv.weight q8_0 +blk.46.ffn_down_shexp.weight q8_0 +blk.46.ffn_gate_shexp.weight q8_0 +blk.46.ffn_up_shexp.weight q8_0 +blk.46.ssm_alpha.weight q8_0 +blk.46.ssm_beta.weight q8_0 +blk.46.ssm_out.weight q8_0 +blk.47.attn_k.weight q8_0 +blk.47.attn_output.weight q8_0 +blk.47.attn_q.weight q8_0 +blk.47.attn_v.weight q8_0 +blk.47.ffn_down_shexp.weight q8_0 +blk.47.ffn_gate_shexp.weight q8_0 +blk.47.ffn_up_shexp.weight q8_0 +blk.48.attn_gate.weight q8_0 +blk.48.attn_qkv.weight q8_0 +blk.48.ffn_down_shexp.weight q8_0 +blk.48.ffn_gate_shexp.weight q8_0 +blk.48.ffn_up_shexp.weight q8_0 +blk.48.ssm_alpha.weight q8_0 +blk.48.ssm_beta.weight q8_0 +blk.48.ssm_out.weight q8_0 +blk.49.attn_gate.weight q8_0 +blk.49.attn_qkv.weight q8_0 +blk.49.ffn_down_shexp.weight q8_0 +blk.49.ffn_gate_shexp.weight q8_0 +blk.49.ffn_up_shexp.weight q8_0 +blk.49.ssm_alpha.weight q8_0 +blk.49.ssm_beta.weight q8_0 +blk.49.ssm_out.weight q8_0 +blk.50.attn_gate.weight q8_0 +blk.50.attn_qkv.weight q8_0 +blk.50.ffn_down_shexp.weight q8_0 +blk.50.ffn_gate_shexp.weight q8_0 +blk.50.ffn_up_shexp.weight q8_0 +blk.50.ssm_alpha.weight q8_0 +blk.50.ssm_beta.weight q8_0 +blk.50.ssm_out.weight q8_0 +blk.51.attn_k.weight q8_0 +blk.51.attn_output.weight q8_0 +blk.51.attn_q.weight q8_0 +blk.51.attn_v.weight q8_0 +blk.51.ffn_down_shexp.weight q8_0 +blk.51.ffn_gate_shexp.weight q8_0 +blk.51.ffn_up_shexp.weight q8_0 +blk.52.attn_gate.weight q8_0 +blk.52.attn_qkv.weight q8_0 +blk.52.ffn_down_shexp.weight q8_0 +blk.52.ffn_gate_shexp.weight q8_0 +blk.52.ffn_up_shexp.weight q8_0 +blk.52.ssm_alpha.weight q8_0 +blk.52.ssm_beta.weight q8_0 +blk.52.ssm_out.weight q8_0 +blk.53.attn_gate.weight q8_0 +blk.53.attn_qkv.weight q8_0 +blk.53.ffn_down_shexp.weight q8_0 +blk.53.ffn_gate_shexp.weight q8_0 +blk.53.ffn_up_shexp.weight q8_0 +blk.53.ssm_alpha.weight q8_0 +blk.53.ssm_beta.weight q8_0 +blk.53.ssm_out.weight q8_0 +blk.54.attn_gate.weight q8_0 +blk.54.attn_qkv.weight q8_0 +blk.54.ffn_down_shexp.weight q8_0 +blk.54.ffn_gate_shexp.weight q8_0 +blk.54.ffn_up_shexp.weight q8_0 +blk.54.ssm_alpha.weight q8_0 +blk.54.ssm_beta.weight q8_0 +blk.54.ssm_out.weight q8_0 +blk.55.attn_k.weight q8_0 +blk.55.attn_output.weight q8_0 +blk.55.attn_q.weight q8_0 +blk.55.attn_v.weight q8_0 +blk.55.ffn_down_shexp.weight q8_0 +blk.55.ffn_gate_shexp.weight q8_0 +blk.55.ffn_up_shexp.weight q8_0 +blk.56.attn_gate.weight q8_0 +blk.56.attn_qkv.weight q8_0 +blk.56.ffn_down_shexp.weight q8_0 +blk.56.ffn_gate_shexp.weight q8_0 +blk.56.ffn_up_shexp.weight q8_0 +blk.56.ssm_alpha.weight q8_0 +blk.56.ssm_beta.weight q8_0 +blk.56.ssm_out.weight q8_0 +blk.57.attn_gate.weight q8_0 +blk.57.attn_qkv.weight q8_0 +blk.57.ffn_down_shexp.weight q8_0 +blk.57.ffn_gate_shexp.weight q8_0 +blk.57.ffn_up_shexp.weight q8_0 +blk.57.ssm_alpha.weight q8_0 +blk.57.ssm_beta.weight q8_0 +blk.57.ssm_out.weight q8_0 +blk.58.attn_gate.weight q8_0 +blk.58.attn_qkv.weight q8_0 +blk.58.ffn_down_shexp.weight q8_0 +blk.58.ffn_gate_shexp.weight q8_0 +blk.58.ffn_up_shexp.weight q8_0 +blk.58.ssm_alpha.weight q8_0 +blk.58.ssm_beta.weight q8_0 +blk.58.ssm_out.weight q8_0 +blk.59.attn_k.weight q8_0 +blk.59.attn_output.weight q8_0 +blk.59.attn_q.weight q8_0 +blk.59.attn_v.weight q8_0 +blk.59.ffn_down_shexp.weight q8_0 +blk.59.ffn_gate_shexp.weight q8_0 +blk.59.ffn_up_shexp.weight q8_0 diff --git a/tests/snapshots/step-3.5-flash.schema b/tests/snapshots/step-3.5-flash.schema new file mode 100644 index 0000000000..c9d1be0082 --- /dev/null +++ b/tests/snapshots/step-3.5-flash.schema @@ -0,0 +1,2453 @@ +# Model: Step-3.5-Flash +# n_embd=4096, n_ff=11264, n_vocab=128896, n_layer=45, n_head=64, n_head_kv=8, n_expert=288 + +[F32] f32 +output.weight q6_K + +[F16] f16 +output.weight q6_K + +[Q4_0] q4_0 +output.weight q6_K + +[Q4_1] q4_1 +output.weight q6_K + +[Q8_0] q8_0 + +[Q5_0] q5_0 +output.weight q6_K + +[Q5_1] q5_1 +output.weight q6_K + +[Q2_K] q2_K +output.weight q6_K +blk.0.attn_output.weight q3_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q3_K +blk.1.attn_output.weight q3_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q3_K +blk.2.attn_output.weight q3_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q3_K +blk.3.attn_output.weight q3_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K +blk.4.attn_output.weight q3_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q3_K +blk.4.ffn_down_shexp.weight q3_K +blk.5.attn_output.weight q3_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K +blk.6.attn_output.weight q3_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.attn_output.weight q3_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.attn_output.weight q3_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.attn_output.weight q3_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_output.weight q3_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.attn_output.weight q3_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.attn_output.weight q3_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.attn_output.weight q3_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_output.weight q3_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.attn_output.weight q3_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.attn_output.weight q3_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.attn_output.weight q3_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_output.weight q3_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.attn_output.weight q3_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.attn_output.weight q3_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.attn_output.weight q3_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_output.weight q3_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.attn_output.weight q3_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.attn_output.weight q3_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.attn_output.weight q3_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_output.weight q3_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.attn_output.weight q3_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.attn_output.weight q3_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.attn_output.weight q3_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_output.weight q3_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.attn_output.weight q3_K +blk.36.attn_v.weight q4_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.attn_output.weight q3_K +blk.37.attn_v.weight q4_K +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.attn_output.weight q3_K +blk.38.attn_v.weight q4_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_output.weight q3_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.attn_output.weight q3_K +blk.40.attn_v.weight q4_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.attn_output.weight q3_K +blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.attn_output.weight q3_K +blk.42.attn_v.weight q4_K +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_output.weight q3_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.attn_output.weight q3_K +blk.44.attn_v.weight q4_K +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K + +[Q3_K_S] q3_K +output.weight q6_K + +[Q3_K_M] q3_K +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q4_K +blk.6.ffn_down_shexp.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q4_K +blk.7.ffn_down_shexp.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q4_K +blk.8.ffn_down_shexp.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q4_K +blk.9.ffn_down_shexp.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q4_K +blk.10.ffn_down_shexp.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q4_K +blk.11.ffn_down_shexp.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q4_K +blk.12.ffn_down_shexp.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q4_K +blk.13.ffn_down_shexp.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q4_K +blk.14.ffn_down_shexp.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q4_K +blk.16.ffn_down_shexp.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q4_K +blk.17.ffn_down_shexp.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q4_K +blk.18.ffn_down_shexp.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q4_K +blk.19.ffn_down_shexp.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q4_K +blk.20.ffn_down_shexp.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q4_K +blk.21.ffn_down_shexp.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q4_K +blk.22.ffn_down_shexp.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.23.ffn_down_shexp.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q4_K +blk.24.ffn_down_shexp.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q4_K +blk.25.ffn_down_shexp.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q4_K +blk.26.ffn_down_shexp.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q4_K +blk.28.ffn_down_shexp.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q4_K +blk.29.ffn_down_shexp.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q4_K +blk.30.ffn_down_shexp.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q4_K +blk.32.ffn_down_shexp.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q4_K +blk.33.ffn_down_shexp.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q4_K +blk.34.ffn_down_shexp.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q4_K +blk.35.ffn_down_shexp.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q4_K +blk.36.ffn_down_exps.weight q4_K +blk.36.ffn_down_shexp.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q4_K +blk.37.ffn_down_exps.weight q4_K +blk.37.ffn_down_shexp.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q4_K +blk.38.ffn_down_exps.weight q4_K +blk.38.ffn_down_shexp.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q4_K +blk.39.ffn_down_shexp.weight q4_K +blk.40.attn_output.weight q4_K +blk.40.attn_v.weight q4_K +blk.40.ffn_down_exps.weight q4_K +blk.40.ffn_down_shexp.weight q4_K +blk.41.attn_output.weight q4_K +blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight q4_K +blk.41.ffn_down_shexp.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K +blk.42.ffn_down_exps.weight q4_K +blk.42.ffn_down_shexp.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K +blk.44.attn_output.weight q4_K +blk.44.attn_v.weight q4_K +blk.44.ffn_down_exps.weight q4_K +blk.44.ffn_down_shexp.weight q4_K + +[Q3_K_L] q3_K +output.weight q6_K +blk.0.attn_output.weight q5_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_output.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.6.ffn_down_exps.weight q5_K +blk.6.ffn_down_shexp.weight q5_K +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q5_K +blk.7.ffn_down_shexp.weight q5_K +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.8.ffn_down_exps.weight q5_K +blk.8.ffn_down_shexp.weight q5_K +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.9.ffn_down_exps.weight q5_K +blk.9.ffn_down_shexp.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.10.ffn_down_exps.weight q5_K +blk.10.ffn_down_shexp.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight q5_K +blk.11.ffn_down_shexp.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down_exps.weight q5_K +blk.12.ffn_down_shexp.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.13.ffn_down_exps.weight q5_K +blk.13.ffn_down_shexp.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.14.ffn_down_exps.weight q5_K +blk.14.ffn_down_shexp.weight q5_K +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.16.ffn_down_exps.weight q5_K +blk.16.ffn_down_shexp.weight q5_K +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.17.ffn_down_exps.weight q5_K +blk.17.ffn_down_shexp.weight q5_K +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down_exps.weight q5_K +blk.18.ffn_down_shexp.weight q5_K +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight q5_K +blk.19.ffn_down_shexp.weight q5_K +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.20.ffn_down_exps.weight q5_K +blk.20.ffn_down_shexp.weight q5_K +blk.21.attn_output.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down_exps.weight q5_K +blk.21.ffn_down_shexp.weight q5_K +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.22.ffn_down_exps.weight q5_K +blk.22.ffn_down_shexp.weight q5_K +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.23.ffn_down_shexp.weight q5_K +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.24.ffn_down_exps.weight q5_K +blk.24.ffn_down_shexp.weight q5_K +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.25.ffn_down_exps.weight q5_K +blk.25.ffn_down_shexp.weight q5_K +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down_exps.weight q5_K +blk.26.ffn_down_shexp.weight q5_K +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.28.ffn_down_exps.weight q5_K +blk.28.ffn_down_shexp.weight q5_K +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down_exps.weight q5_K +blk.29.ffn_down_shexp.weight q5_K +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.30.ffn_down_exps.weight q5_K +blk.30.ffn_down_shexp.weight q5_K +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down_exps.weight q5_K +blk.32.ffn_down_shexp.weight q5_K +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.33.ffn_down_exps.weight q5_K +blk.33.ffn_down_shexp.weight q5_K +blk.34.attn_output.weight q5_K +blk.34.attn_v.weight q5_K +blk.34.ffn_down_exps.weight q5_K +blk.34.ffn_down_shexp.weight q5_K +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight q5_K +blk.35.ffn_down_shexp.weight q5_K +blk.36.attn_output.weight q5_K +blk.36.attn_v.weight q5_K +blk.36.ffn_down_exps.weight q5_K +blk.36.ffn_down_shexp.weight q5_K +blk.37.attn_output.weight q5_K +blk.37.attn_v.weight q5_K +blk.37.ffn_down_exps.weight q5_K +blk.37.ffn_down_shexp.weight q5_K +blk.38.attn_output.weight q5_K +blk.38.attn_v.weight q5_K +blk.38.ffn_down_exps.weight q5_K +blk.38.ffn_down_shexp.weight q5_K +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down_exps.weight q5_K +blk.39.ffn_down_shexp.weight q5_K +blk.40.attn_output.weight q5_K +blk.40.attn_v.weight q5_K +blk.40.ffn_down_exps.weight q5_K +blk.40.ffn_down_shexp.weight q5_K +blk.41.attn_output.weight q5_K +blk.41.attn_v.weight q5_K +blk.41.ffn_down_exps.weight q5_K +blk.41.ffn_down_shexp.weight q5_K +blk.42.attn_output.weight q5_K +blk.42.attn_v.weight q5_K +blk.42.ffn_down_exps.weight q5_K +blk.42.ffn_down_shexp.weight q5_K +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K +blk.44.attn_output.weight q5_K +blk.44.attn_v.weight q5_K +blk.44.ffn_down_exps.weight q5_K +blk.44.ffn_down_shexp.weight q5_K + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K + +[Q4_K_M] q4_K +output.weight q6_K +blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down_exps.weight q6_K +blk.7.ffn_down_shexp.weight q6_K +blk.10.attn_v.weight q6_K +blk.10.ffn_down_exps.weight q6_K +blk.10.ffn_down_shexp.weight q6_K +blk.13.attn_v.weight q6_K +blk.13.ffn_down_exps.weight q6_K +blk.13.ffn_down_shexp.weight q6_K +blk.16.attn_v.weight q6_K +blk.16.ffn_down_exps.weight q6_K +blk.16.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down_exps.weight q6_K +blk.19.ffn_down_shexp.weight q6_K +blk.22.attn_v.weight q6_K +blk.22.ffn_down_exps.weight q6_K +blk.22.ffn_down_shexp.weight q6_K +blk.25.attn_v.weight q6_K +blk.25.ffn_down_exps.weight q6_K +blk.25.ffn_down_shexp.weight q6_K +blk.28.attn_v.weight q6_K +blk.28.ffn_down_exps.weight q6_K +blk.28.ffn_down_shexp.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down_exps.weight q6_K +blk.31.ffn_down_shexp.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down_exps.weight q6_K +blk.34.ffn_down_shexp.weight q6_K +blk.37.attn_v.weight q6_K +blk.37.ffn_down_exps.weight q6_K +blk.37.ffn_down_shexp.weight q6_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.40.attn_v.weight q6_K +blk.40.ffn_down_exps.weight q6_K +blk.40.ffn_down_shexp.weight q6_K +blk.41.attn_v.weight q6_K +blk.41.ffn_down_exps.weight q6_K +blk.41.ffn_down_shexp.weight q6_K +blk.42.attn_v.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.43.attn_v.weight q6_K +blk.43.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K +blk.44.attn_v.weight q6_K +blk.44.ffn_down_exps.weight q6_K +blk.44.ffn_down_shexp.weight q6_K + +[Q5_K_S] q5_K +output.weight q6_K + +[Q5_K_M] q5_K +output.weight q6_K +blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K +blk.1.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K +blk.2.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K +blk.4.attn_v.weight q6_K +blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K +blk.7.attn_v.weight q6_K +blk.7.ffn_down_exps.weight q6_K +blk.7.ffn_down_shexp.weight q6_K +blk.10.attn_v.weight q6_K +blk.10.ffn_down_exps.weight q6_K +blk.10.ffn_down_shexp.weight q6_K +blk.13.attn_v.weight q6_K +blk.13.ffn_down_exps.weight q6_K +blk.13.ffn_down_shexp.weight q6_K +blk.16.attn_v.weight q6_K +blk.16.ffn_down_exps.weight q6_K +blk.16.ffn_down_shexp.weight q6_K +blk.19.attn_v.weight q6_K +blk.19.ffn_down_exps.weight q6_K +blk.19.ffn_down_shexp.weight q6_K +blk.22.attn_v.weight q6_K +blk.22.ffn_down_exps.weight q6_K +blk.22.ffn_down_shexp.weight q6_K +blk.25.attn_v.weight q6_K +blk.25.ffn_down_exps.weight q6_K +blk.25.ffn_down_shexp.weight q6_K +blk.28.attn_v.weight q6_K +blk.28.ffn_down_exps.weight q6_K +blk.28.ffn_down_shexp.weight q6_K +blk.31.attn_v.weight q6_K +blk.31.ffn_down_exps.weight q6_K +blk.31.ffn_down_shexp.weight q6_K +blk.34.attn_v.weight q6_K +blk.34.ffn_down_exps.weight q6_K +blk.34.ffn_down_shexp.weight q6_K +blk.37.attn_v.weight q6_K +blk.37.ffn_down_exps.weight q6_K +blk.37.ffn_down_shexp.weight q6_K +blk.39.attn_v.weight q6_K +blk.39.ffn_down_exps.weight q6_K +blk.39.ffn_down_shexp.weight q6_K +blk.40.attn_v.weight q6_K +blk.40.ffn_down_exps.weight q6_K +blk.40.ffn_down_shexp.weight q6_K +blk.41.attn_v.weight q6_K +blk.41.ffn_down_exps.weight q6_K +blk.41.ffn_down_shexp.weight q6_K +blk.42.attn_v.weight q6_K +blk.42.ffn_down_exps.weight q6_K +blk.42.ffn_down_shexp.weight q6_K +blk.43.attn_v.weight q6_K +blk.43.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K +blk.44.attn_v.weight q6_K +blk.44.ffn_down_exps.weight q6_K +blk.44.ffn_down_shexp.weight q6_K + +[Q6_K] q6_K + +[IQ2_XXS] iq2_xxs +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K + +[IQ2_XS] iq2_xs +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K + +[Q2_K_S] q2_K +output.weight q6_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K + +[IQ3_XS] iq3_s +output.weight q6_K +blk.0.attn_k.weight iq3_xxs +blk.0.attn_q.weight iq3_xxs +blk.0.attn_v.weight q4_K +blk.1.attn_k.weight iq3_xxs +blk.1.attn_q.weight iq3_xxs +blk.1.attn_v.weight q4_K +blk.2.attn_k.weight iq3_xxs +blk.2.attn_q.weight iq3_xxs +blk.2.attn_v.weight q4_K +blk.3.attn_k.weight iq3_xxs +blk.3.attn_q.weight iq3_xxs +blk.3.attn_v.weight q4_K +blk.4.attn_k.weight iq3_xxs +blk.4.attn_q.weight iq3_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_k.weight iq3_xxs +blk.5.attn_q.weight iq3_xxs +blk.5.attn_v.weight q4_K +blk.5.ffn_gate_exps.weight iq3_xxs +blk.5.ffn_gate_shexp.weight iq3_xxs +blk.5.ffn_up_exps.weight iq3_xxs +blk.5.ffn_up_shexp.weight iq3_xxs +blk.6.attn_k.weight iq3_xxs +blk.6.attn_q.weight iq3_xxs +blk.6.attn_v.weight q4_K +blk.6.ffn_gate_exps.weight iq3_xxs +blk.6.ffn_gate_shexp.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs +blk.6.ffn_up_shexp.weight iq3_xxs +blk.7.attn_k.weight iq3_xxs +blk.7.attn_q.weight iq3_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs +blk.8.attn_k.weight iq3_xxs +blk.8.attn_q.weight iq3_xxs +blk.8.attn_v.weight q4_K +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs +blk.9.attn_k.weight iq3_xxs +blk.9.attn_q.weight iq3_xxs +blk.9.attn_v.weight q4_K +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.9.ffn_up_shexp.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.10.attn_v.weight q4_K +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.12.attn_v.weight q4_K +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.12.ffn_up_shexp.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.13.attn_v.weight q4_K +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.14.attn_v.weight q4_K +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs +blk.15.attn_k.weight iq3_xxs +blk.15.attn_q.weight iq3_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs +blk.16.attn_k.weight iq3_xxs +blk.16.attn_q.weight iq3_xxs +blk.16.attn_v.weight q4_K +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs +blk.17.attn_k.weight iq3_xxs +blk.17.attn_q.weight iq3_xxs +blk.17.attn_v.weight q4_K +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs +blk.18.attn_k.weight iq3_xxs +blk.18.attn_q.weight iq3_xxs +blk.18.attn_v.weight q4_K +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs +blk.19.attn_k.weight iq3_xxs +blk.19.attn_q.weight iq3_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs +blk.20.attn_k.weight iq3_xxs +blk.20.attn_q.weight iq3_xxs +blk.20.attn_v.weight q4_K +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs +blk.21.attn_k.weight iq3_xxs +blk.21.attn_q.weight iq3_xxs +blk.21.attn_v.weight q4_K +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs +blk.22.attn_k.weight iq3_xxs +blk.22.attn_q.weight iq3_xxs +blk.22.attn_v.weight q4_K +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs +blk.23.attn_k.weight iq3_xxs +blk.23.attn_q.weight iq3_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs +blk.24.attn_k.weight iq3_xxs +blk.24.attn_q.weight iq3_xxs +blk.24.attn_v.weight q4_K +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_gate_shexp.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs +blk.25.attn_k.weight iq3_xxs +blk.25.attn_q.weight iq3_xxs +blk.25.attn_v.weight q4_K +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs +blk.26.attn_k.weight iq3_xxs +blk.26.attn_q.weight iq3_xxs +blk.26.attn_v.weight q4_K +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs +blk.27.attn_k.weight iq3_xxs +blk.27.attn_q.weight iq3_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs +blk.28.attn_k.weight iq3_xxs +blk.28.attn_q.weight iq3_xxs +blk.28.attn_v.weight q4_K +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs +blk.29.attn_k.weight iq3_xxs +blk.29.attn_q.weight iq3_xxs +blk.29.attn_v.weight q4_K +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs +blk.30.attn_k.weight iq3_xxs +blk.30.attn_q.weight iq3_xxs +blk.30.attn_v.weight q4_K +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs +blk.31.attn_k.weight iq3_xxs +blk.31.attn_q.weight iq3_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs +blk.32.attn_k.weight iq3_xxs +blk.32.attn_q.weight iq3_xxs +blk.32.attn_v.weight q4_K +blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs +blk.32.ffn_up_exps.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs +blk.33.attn_k.weight iq3_xxs +blk.33.attn_q.weight iq3_xxs +blk.33.attn_v.weight q4_K +blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs +blk.33.ffn_up_exps.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs +blk.34.attn_k.weight iq3_xxs +blk.34.attn_q.weight iq3_xxs +blk.34.attn_v.weight q4_K +blk.34.ffn_gate_exps.weight iq3_xxs +blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs +blk.35.attn_k.weight iq3_xxs +blk.35.attn_q.weight iq3_xxs +blk.35.attn_v.weight q4_K +blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs +blk.35.ffn_up_exps.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs +blk.36.attn_k.weight iq3_xxs +blk.36.attn_q.weight iq3_xxs +blk.36.attn_v.weight q4_K +blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs +blk.37.attn_k.weight iq3_xxs +blk.37.attn_q.weight iq3_xxs +blk.37.attn_v.weight q4_K +blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs +blk.37.ffn_up_exps.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs +blk.38.attn_k.weight iq3_xxs +blk.38.attn_q.weight iq3_xxs +blk.38.attn_v.weight q4_K +blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs +blk.39.attn_k.weight iq3_xxs +blk.39.attn_q.weight iq3_xxs +blk.39.attn_v.weight q4_K +blk.40.attn_k.weight iq3_xxs +blk.40.attn_q.weight iq3_xxs +blk.40.attn_v.weight q4_K +blk.41.attn_k.weight iq3_xxs +blk.41.attn_q.weight iq3_xxs +blk.41.attn_v.weight q4_K +blk.42.attn_k.weight iq3_xxs +blk.42.attn_q.weight iq3_xxs +blk.42.attn_v.weight q4_K +blk.43.attn_k.weight iq3_xxs +blk.43.attn_q.weight iq3_xxs +blk.43.attn_v.weight q4_K +blk.44.attn_k.weight iq3_xxs +blk.44.attn_q.weight iq3_xxs +blk.44.attn_v.weight q4_K + +[IQ3_XXS] iq3_xxs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_k.weight iq2_s +blk.0.attn_output.weight iq3_s +blk.0.attn_q.weight iq2_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_k.weight iq2_s +blk.1.attn_output.weight iq3_s +blk.1.attn_q.weight iq2_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_k.weight iq2_s +blk.2.attn_output.weight iq3_s +blk.2.attn_q.weight iq2_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_k.weight iq2_s +blk.3.attn_output.weight iq3_s +blk.3.attn_q.weight iq2_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_k.weight iq2_s +blk.4.attn_output.weight iq3_s +blk.4.attn_q.weight iq2_s +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_k.weight iq2_s +blk.5.attn_output.weight iq3_s +blk.5.attn_q.weight iq2_s +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K +blk.6.attn_k.weight iq2_s +blk.6.attn_output.weight iq3_s +blk.6.attn_q.weight iq2_s +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight q3_K +blk.6.ffn_down_shexp.weight q3_K +blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s +blk.7.attn_q.weight iq2_s +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K +blk.8.attn_k.weight iq2_s +blk.8.attn_output.weight iq3_s +blk.8.attn_q.weight iq2_s +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K +blk.9.attn_k.weight iq2_s +blk.9.attn_output.weight iq3_s +blk.9.attn_q.weight iq2_s +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight q3_K +blk.9.ffn_down_shexp.weight q3_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight q3_K +blk.10.ffn_down_shexp.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight q3_K +blk.12.ffn_down_shexp.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight q3_K +blk.14.ffn_down_shexp.weight q3_K +blk.15.attn_k.weight iq2_s +blk.15.attn_output.weight iq3_s +blk.15.attn_q.weight iq2_s +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K +blk.16.attn_k.weight iq2_s +blk.16.attn_output.weight iq3_s +blk.16.attn_q.weight iq2_s +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K +blk.17.attn_k.weight iq2_s +blk.17.attn_output.weight iq3_s +blk.17.attn_q.weight iq2_s +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight q3_K +blk.17.ffn_down_shexp.weight q3_K +blk.18.attn_k.weight iq2_s +blk.18.attn_output.weight iq3_s +blk.18.attn_q.weight iq2_s +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight q3_K +blk.18.ffn_down_shexp.weight q3_K +blk.19.attn_k.weight iq2_s +blk.19.attn_output.weight iq3_s +blk.19.attn_q.weight iq2_s +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K +blk.20.attn_k.weight iq2_s +blk.20.attn_output.weight iq3_s +blk.20.attn_q.weight iq2_s +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K +blk.21.attn_k.weight iq2_s +blk.21.attn_output.weight iq3_s +blk.21.attn_q.weight iq2_s +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K +blk.22.attn_k.weight iq2_s +blk.22.attn_output.weight iq3_s +blk.22.attn_q.weight iq2_s +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight q3_K +blk.22.ffn_down_shexp.weight q3_K +blk.23.attn_k.weight iq2_s +blk.23.attn_output.weight iq3_s +blk.23.attn_q.weight iq2_s +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K +blk.24.attn_k.weight iq2_s +blk.24.attn_output.weight iq3_s +blk.24.attn_q.weight iq2_s +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K +blk.25.attn_k.weight iq2_s +blk.25.attn_output.weight iq3_s +blk.25.attn_q.weight iq2_s +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K +blk.26.attn_k.weight iq2_s +blk.26.attn_output.weight iq3_s +blk.26.attn_q.weight iq2_s +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight q3_K +blk.26.ffn_down_shexp.weight q3_K +blk.27.attn_k.weight iq2_s +blk.27.attn_output.weight iq3_s +blk.27.attn_q.weight iq2_s +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K +blk.28.attn_k.weight iq2_s +blk.28.attn_output.weight iq3_s +blk.28.attn_q.weight iq2_s +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K +blk.29.attn_k.weight iq2_s +blk.29.attn_output.weight iq3_s +blk.29.attn_q.weight iq2_s +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight q3_K +blk.29.ffn_down_shexp.weight q3_K +blk.30.attn_k.weight iq2_s +blk.30.attn_output.weight iq3_s +blk.30.attn_q.weight iq2_s +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight q3_K +blk.30.ffn_down_shexp.weight q3_K +blk.31.attn_k.weight iq2_s +blk.31.attn_output.weight iq3_s +blk.31.attn_q.weight iq2_s +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K +blk.32.attn_k.weight iq2_s +blk.32.attn_output.weight iq3_s +blk.32.attn_q.weight iq2_s +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K +blk.33.attn_k.weight iq2_s +blk.33.attn_output.weight iq3_s +blk.33.attn_q.weight iq2_s +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K +blk.34.attn_k.weight iq2_s +blk.34.attn_output.weight iq3_s +blk.34.attn_q.weight iq2_s +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight q3_K +blk.34.ffn_down_shexp.weight q3_K +blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s +blk.35.attn_q.weight iq2_s +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K +blk.36.attn_k.weight iq2_s +blk.36.attn_output.weight iq3_s +blk.36.attn_q.weight iq2_s +blk.36.attn_v.weight q4_K +blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K +blk.37.attn_k.weight iq2_s +blk.37.attn_output.weight iq3_s +blk.37.attn_q.weight iq2_s +blk.37.attn_v.weight q4_K +blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K +blk.38.attn_k.weight iq2_s +blk.38.attn_output.weight iq3_s +blk.38.attn_q.weight iq2_s +blk.38.attn_v.weight q4_K +blk.38.ffn_down_exps.weight q3_K +blk.38.ffn_down_shexp.weight q3_K +blk.39.attn_k.weight iq2_s +blk.39.attn_output.weight iq3_s +blk.39.attn_q.weight iq2_s +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K +blk.40.attn_k.weight iq2_s +blk.40.attn_output.weight iq3_s +blk.40.attn_q.weight iq2_s +blk.40.attn_v.weight q4_K +blk.40.ffn_down_exps.weight q3_K +blk.40.ffn_down_shexp.weight q3_K +blk.41.attn_k.weight iq2_s +blk.41.attn_output.weight iq3_s +blk.41.attn_q.weight iq2_s +blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K +blk.42.attn_k.weight iq2_s +blk.42.attn_output.weight iq3_s +blk.42.attn_q.weight iq2_s +blk.42.attn_v.weight q4_K +blk.42.ffn_down_exps.weight q3_K +blk.42.ffn_down_shexp.weight q3_K +blk.43.attn_k.weight iq2_s +blk.43.attn_output.weight iq3_s +blk.43.attn_q.weight iq2_s +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K +blk.44.attn_k.weight iq2_s +blk.44.attn_output.weight iq3_s +blk.44.attn_q.weight iq2_s +blk.44.attn_v.weight q4_K +blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K + +[IQ1_S] iq1_s +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq2_xxs +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq2_xxs +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq2_xxs +blk.44.attn_v.weight q4_K + +[IQ4_NL] iq4_nl +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K + +[IQ3_S] iq3_s +output.weight q6_K +blk.0.attn_v.weight q4_K +blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K + +[IQ3_M] iq3_s +output.weight q6_K +blk.0.attn_output.weight q4_K +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_K +blk.1.attn_output.weight q4_K +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K +blk.2.attn_output.weight q4_K +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.attn_output.weight q4_K +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K +blk.4.attn_output.weight q4_K +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K +blk.5.attn_output.weight q4_K +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight q4_K +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight q4_K +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight q4_K +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight q4_K +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight q4_K +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight q4_K +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight q4_K +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight q4_K +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight q4_K +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight q4_K +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight q4_K +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight q4_K +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight q4_K +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight q4_K +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight q4_K +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight q4_K +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight q4_K +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight q4_K +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight q4_K +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight q4_K +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight q4_K +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight q4_K +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight q4_K +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight q4_K +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight q4_K +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight q4_K +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight q4_K +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight q4_K +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight q4_K +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight q4_K +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight q4_K +blk.44.attn_v.weight q4_K + +[IQ2_S] iq2_xs +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq3_s +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq3_s +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq3_s +blk.44.attn_v.weight q4_K + +[IQ2_M] iq2_s +output.weight q5_K +token_embd.weight iq3_s +blk.0.attn_output.weight iq3_s +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s +blk.1.attn_output.weight iq3_s +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight iq3_s +blk.2.attn_output.weight iq3_s +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight iq3_s +blk.3.attn_output.weight iq3_s +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq3_s +blk.3.ffn_down_shexp.weight iq3_s +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq3_s +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq3_s +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq3_s +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq3_s +blk.44.attn_v.weight q4_K + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K +blk.1.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K +blk.5.attn_v.weight q5_K +blk.6.attn_v.weight q5_K +blk.7.attn_v.weight q5_K +blk.8.attn_v.weight q5_K +blk.9.attn_v.weight q5_K +blk.10.attn_v.weight q5_K +blk.11.attn_v.weight q5_K +blk.12.attn_v.weight q5_K +blk.13.attn_v.weight q5_K +blk.14.attn_v.weight q5_K +blk.15.attn_v.weight q5_K +blk.16.attn_v.weight q5_K +blk.17.attn_v.weight q5_K +blk.18.attn_v.weight q5_K +blk.19.attn_v.weight q5_K +blk.20.attn_v.weight q5_K +blk.21.attn_v.weight q5_K +blk.22.attn_v.weight q5_K +blk.23.attn_v.weight q5_K +blk.24.attn_v.weight q5_K +blk.25.attn_v.weight q5_K +blk.26.attn_v.weight q5_K +blk.27.attn_v.weight q5_K +blk.28.attn_v.weight q5_K +blk.29.attn_v.weight q5_K +blk.30.attn_v.weight q5_K +blk.31.attn_v.weight q5_K +blk.32.attn_v.weight q5_K +blk.33.attn_v.weight q5_K +blk.34.attn_v.weight q5_K +blk.35.attn_v.weight q5_K +blk.36.attn_v.weight q5_K +blk.37.attn_v.weight q5_K +blk.38.attn_v.weight q5_K +blk.39.attn_v.weight q5_K +blk.40.attn_v.weight q5_K +blk.41.attn_v.weight q5_K +blk.42.attn_v.weight q5_K +blk.43.attn_v.weight q5_K +blk.44.attn_v.weight q5_K + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down.weight q2_K +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q2_K +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight q2_K +blk.3.ffn_down_shexp.weight q2_K +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq2_xxs +blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq2_xxs +blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq2_xxs +blk.44.attn_v.weight q4_K + +[BF16] bf16 +output.weight q6_K + +[TQ1_0] tq1_0 +output.weight q6_K +token_embd.weight q4_K + +[TQ2_0] tq2_0 +output.weight q6_K +token_embd.weight q4_K + +[MXFP4_MOE] mxfp4 +output.weight q8_0 +token_embd.weight q8_0 +blk.0.attn_gate.weight q8_0 +blk.0.attn_k.weight q8_0 +blk.0.attn_output.weight q8_0 +blk.0.attn_q.weight q8_0 +blk.0.attn_v.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 +blk.1.attn_gate.weight q8_0 +blk.1.attn_k.weight q8_0 +blk.1.attn_output.weight q8_0 +blk.1.attn_q.weight q8_0 +blk.1.attn_v.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 +blk.2.attn_gate.weight q8_0 +blk.2.attn_k.weight q8_0 +blk.2.attn_output.weight q8_0 +blk.2.attn_q.weight q8_0 +blk.2.attn_v.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 +blk.3.attn_gate.weight q8_0 +blk.3.attn_k.weight q8_0 +blk.3.attn_output.weight q8_0 +blk.3.attn_q.weight q8_0 +blk.3.attn_v.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 +blk.4.attn_gate.weight q8_0 +blk.4.attn_k.weight q8_0 +blk.4.attn_output.weight q8_0 +blk.4.attn_q.weight q8_0 +blk.4.attn_v.weight q8_0 +blk.4.ffn_down_shexp.weight q8_0 +blk.4.ffn_gate_shexp.weight q8_0 +blk.4.ffn_up_shexp.weight q8_0 +blk.5.attn_gate.weight q8_0 +blk.5.attn_k.weight q8_0 +blk.5.attn_output.weight q8_0 +blk.5.attn_q.weight q8_0 +blk.5.attn_v.weight q8_0 +blk.5.ffn_down_shexp.weight q8_0 +blk.5.ffn_gate_shexp.weight q8_0 +blk.5.ffn_up_shexp.weight q8_0 +blk.6.attn_gate.weight q8_0 +blk.6.attn_k.weight q8_0 +blk.6.attn_output.weight q8_0 +blk.6.attn_q.weight q8_0 +blk.6.attn_v.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_gate_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 +blk.7.attn_gate.weight q8_0 +blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 +blk.7.attn_q.weight q8_0 +blk.7.attn_v.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 +blk.8.attn_gate.weight q8_0 +blk.8.attn_k.weight q8_0 +blk.8.attn_output.weight q8_0 +blk.8.attn_q.weight q8_0 +blk.8.attn_v.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_gate_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 +blk.9.attn_gate.weight q8_0 +blk.9.attn_k.weight q8_0 +blk.9.attn_output.weight q8_0 +blk.9.attn_q.weight q8_0 +blk.9.attn_v.weight q8_0 +blk.9.ffn_down_shexp.weight q8_0 +blk.9.ffn_gate_shexp.weight q8_0 +blk.9.ffn_up_shexp.weight q8_0 +blk.10.attn_gate.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_gate_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 +blk.11.attn_gate.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 +blk.12.attn_gate.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.12.ffn_down_shexp.weight q8_0 +blk.12.ffn_gate_shexp.weight q8_0 +blk.12.ffn_up_shexp.weight q8_0 +blk.13.attn_gate.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_gate_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 +blk.14.attn_gate.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 +blk.14.ffn_down_shexp.weight q8_0 +blk.14.ffn_gate_shexp.weight q8_0 +blk.14.ffn_up_shexp.weight q8_0 +blk.15.attn_gate.weight q8_0 +blk.15.attn_k.weight q8_0 +blk.15.attn_output.weight q8_0 +blk.15.attn_q.weight q8_0 +blk.15.attn_v.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 +blk.16.attn_gate.weight q8_0 +blk.16.attn_k.weight q8_0 +blk.16.attn_output.weight q8_0 +blk.16.attn_q.weight q8_0 +blk.16.attn_v.weight q8_0 +blk.16.ffn_down_shexp.weight q8_0 +blk.16.ffn_gate_shexp.weight q8_0 +blk.16.ffn_up_shexp.weight q8_0 +blk.17.attn_gate.weight q8_0 +blk.17.attn_k.weight q8_0 +blk.17.attn_output.weight q8_0 +blk.17.attn_q.weight q8_0 +blk.17.attn_v.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_gate_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 +blk.18.attn_gate.weight q8_0 +blk.18.attn_k.weight q8_0 +blk.18.attn_output.weight q8_0 +blk.18.attn_q.weight q8_0 +blk.18.attn_v.weight q8_0 +blk.18.ffn_down_shexp.weight q8_0 +blk.18.ffn_gate_shexp.weight q8_0 +blk.18.ffn_up_shexp.weight q8_0 +blk.19.attn_gate.weight q8_0 +blk.19.attn_k.weight q8_0 +blk.19.attn_output.weight q8_0 +blk.19.attn_q.weight q8_0 +blk.19.attn_v.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 +blk.20.attn_gate.weight q8_0 +blk.20.attn_k.weight q8_0 +blk.20.attn_output.weight q8_0 +blk.20.attn_q.weight q8_0 +blk.20.attn_v.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_gate_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 +blk.21.attn_gate.weight q8_0 +blk.21.attn_k.weight q8_0 +blk.21.attn_output.weight q8_0 +blk.21.attn_q.weight q8_0 +blk.21.attn_v.weight q8_0 +blk.21.ffn_down_shexp.weight q8_0 +blk.21.ffn_gate_shexp.weight q8_0 +blk.21.ffn_up_shexp.weight q8_0 +blk.22.attn_gate.weight q8_0 +blk.22.attn_k.weight q8_0 +blk.22.attn_output.weight q8_0 +blk.22.attn_q.weight q8_0 +blk.22.attn_v.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_gate_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 +blk.23.attn_gate.weight q8_0 +blk.23.attn_k.weight q8_0 +blk.23.attn_output.weight q8_0 +blk.23.attn_q.weight q8_0 +blk.23.attn_v.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 +blk.24.attn_gate.weight q8_0 +blk.24.attn_k.weight q8_0 +blk.24.attn_output.weight q8_0 +blk.24.attn_q.weight q8_0 +blk.24.attn_v.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_gate_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 +blk.25.attn_gate.weight q8_0 +blk.25.attn_k.weight q8_0 +blk.25.attn_output.weight q8_0 +blk.25.attn_q.weight q8_0 +blk.25.attn_v.weight q8_0 +blk.25.ffn_down_shexp.weight q8_0 +blk.25.ffn_gate_shexp.weight q8_0 +blk.25.ffn_up_shexp.weight q8_0 +blk.26.attn_gate.weight q8_0 +blk.26.attn_k.weight q8_0 +blk.26.attn_output.weight q8_0 +blk.26.attn_q.weight q8_0 +blk.26.attn_v.weight q8_0 +blk.26.ffn_down_shexp.weight q8_0 +blk.26.ffn_gate_shexp.weight q8_0 +blk.26.ffn_up_shexp.weight q8_0 +blk.27.attn_gate.weight q8_0 +blk.27.attn_k.weight q8_0 +blk.27.attn_output.weight q8_0 +blk.27.attn_q.weight q8_0 +blk.27.attn_v.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 +blk.28.attn_gate.weight q8_0 +blk.28.attn_k.weight q8_0 +blk.28.attn_output.weight q8_0 +blk.28.attn_q.weight q8_0 +blk.28.attn_v.weight q8_0 +blk.28.ffn_down_shexp.weight q8_0 +blk.28.ffn_gate_shexp.weight q8_0 +blk.28.ffn_up_shexp.weight q8_0 +blk.29.attn_gate.weight q8_0 +blk.29.attn_k.weight q8_0 +blk.29.attn_output.weight q8_0 +blk.29.attn_q.weight q8_0 +blk.29.attn_v.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_gate_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 +blk.30.attn_gate.weight q8_0 +blk.30.attn_k.weight q8_0 +blk.30.attn_output.weight q8_0 +blk.30.attn_q.weight q8_0 +blk.30.attn_v.weight q8_0 +blk.30.ffn_down_shexp.weight q8_0 +blk.30.ffn_gate_shexp.weight q8_0 +blk.30.ffn_up_shexp.weight q8_0 +blk.31.attn_gate.weight q8_0 +blk.31.attn_k.weight q8_0 +blk.31.attn_output.weight q8_0 +blk.31.attn_q.weight q8_0 +blk.31.attn_v.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 +blk.32.attn_gate.weight q8_0 +blk.32.attn_k.weight q8_0 +blk.32.attn_output.weight q8_0 +blk.32.attn_q.weight q8_0 +blk.32.attn_v.weight q8_0 +blk.32.ffn_down_shexp.weight q8_0 +blk.32.ffn_gate_shexp.weight q8_0 +blk.32.ffn_up_shexp.weight q8_0 +blk.33.attn_gate.weight q8_0 +blk.33.attn_k.weight q8_0 +blk.33.attn_output.weight q8_0 +blk.33.attn_q.weight q8_0 +blk.33.attn_v.weight q8_0 +blk.33.ffn_down_shexp.weight q8_0 +blk.33.ffn_gate_shexp.weight q8_0 +blk.33.ffn_up_shexp.weight q8_0 +blk.34.attn_gate.weight q8_0 +blk.34.attn_k.weight q8_0 +blk.34.attn_output.weight q8_0 +blk.34.attn_q.weight q8_0 +blk.34.attn_v.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_gate_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 +blk.35.attn_gate.weight q8_0 +blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 +blk.35.attn_q.weight q8_0 +blk.35.attn_v.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 +blk.36.attn_gate.weight q8_0 +blk.36.attn_k.weight q8_0 +blk.36.attn_output.weight q8_0 +blk.36.attn_q.weight q8_0 +blk.36.attn_v.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_gate_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 +blk.37.attn_gate.weight q8_0 +blk.37.attn_k.weight q8_0 +blk.37.attn_output.weight q8_0 +blk.37.attn_q.weight q8_0 +blk.37.attn_v.weight q8_0 +blk.37.ffn_down_shexp.weight q8_0 +blk.37.ffn_gate_shexp.weight q8_0 +blk.37.ffn_up_shexp.weight q8_0 +blk.38.attn_gate.weight q8_0 +blk.38.attn_k.weight q8_0 +blk.38.attn_output.weight q8_0 +blk.38.attn_q.weight q8_0 +blk.38.attn_v.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_gate_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 +blk.39.attn_gate.weight q8_0 +blk.39.attn_k.weight q8_0 +blk.39.attn_output.weight q8_0 +blk.39.attn_q.weight q8_0 +blk.39.attn_v.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 +blk.40.attn_gate.weight q8_0 +blk.40.attn_k.weight q8_0 +blk.40.attn_output.weight q8_0 +blk.40.attn_q.weight q8_0 +blk.40.attn_v.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_gate_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 +blk.41.attn_gate.weight q8_0 +blk.41.attn_k.weight q8_0 +blk.41.attn_output.weight q8_0 +blk.41.attn_q.weight q8_0 +blk.41.attn_v.weight q8_0 +blk.41.ffn_down_shexp.weight q8_0 +blk.41.ffn_gate_shexp.weight q8_0 +blk.41.ffn_up_shexp.weight q8_0 +blk.42.attn_gate.weight q8_0 +blk.42.attn_k.weight q8_0 +blk.42.attn_output.weight q8_0 +blk.42.attn_q.weight q8_0 +blk.42.attn_v.weight q8_0 +blk.42.ffn_down_shexp.weight q8_0 +blk.42.ffn_gate_shexp.weight q8_0 +blk.42.ffn_up_shexp.weight q8_0 +blk.43.attn_gate.weight q8_0 +blk.43.attn_k.weight q8_0 +blk.43.attn_output.weight q8_0 +blk.43.attn_q.weight q8_0 +blk.43.attn_v.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 +blk.44.attn_gate.weight q8_0 +blk.44.attn_k.weight q8_0 +blk.44.attn_output.weight q8_0 +blk.44.attn_q.weight q8_0 +blk.44.attn_v.weight q8_0 +blk.44.ffn_down_shexp.weight q8_0 +blk.44.ffn_gate_shexp.weight q8_0 +blk.44.ffn_up_shexp.weight q8_0 diff --git a/tests/test-quant-type-selection.cpp b/tests/test-quant-type-selection.cpp index 4ecb4c09ac..26d04f7c38 100644 --- a/tests/test-quant-type-selection.cpp +++ b/tests/test-quant-type-selection.cpp @@ -190,7 +190,7 @@ static const remote_model_spec model_specs[] = { { "ggml-org/gpt-oss-120b-GGUF", "mxfp4" }, { "ggml-org/gemma-3-4b-it-GGUF", "Q8_0" }, { "bartowski/Meta-Llama-3.1-70B-Instruct-GGUF", "Q4_K_M" }, - { "bartowski/deepseek-ai_DeepSeek-V3.1-GGUF", "IQ1_S" }, + { "bartowski/deepseek-ai_DeepSeek-V3.1-GGUF", "IQ1_M" }, { "bartowski/Qwen_Qwen3.5-397B-A17B-GGUF", "IQ1_S" }, // TODO: swap with ggml-org if/when it's released { "bartowski/Qwen_Qwen3.5-27B-GGUF", "Q8_0" }, // TODO: swap with ggml-org if/when it's released };