happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

memory: respect unified KV cache in hybrid memory for eval tasks (#21224) 

The hybrid memory paths (`llama-memory-hybrid.cpp` and
`llama-memory-hybrid-iswa.cpp`) always used sequential equal split,
ignoring the unified KV cache flag. This caused hellaswag, winogrande,
and multiple-choice evaluations to fail on hybrid models (models with
both attention and recurrent/SSM layers, such as Qwen3.5-35B-A3B) with:

  split_equal: sequential split is not supported when there are
  coupled sequences in the input batch (you may need to use the
  -kvu flag)

PR #19954 fixed this for `llama-kv-cache-iswa.cpp` by automatically
enabling unified KV mode and setting n_parallel >= 4 for multi-choice
eval tasks. However, the hybrid memory paths were not updated.

This commit mirrors the iswa fix: use non-sequential split when KV
cache is unified (n_stream == 1), which is automatically set by
llama-perplexity for hellaswag/winogrande/multiple-choice since #19954.

Tested on Qwen3.5-35B-A3B (hybrid attention+SSM MoE model):
- HellaSwag: 83.0% (400 tasks)
- Winogrande: 74.5% (400 tasks)
- MMLU: 41.2%
- ARC-Challenge: 56.2%
- TruthfulQA: 37.7%
All previously failed with llama_decode() error.
This commit is contained in:
Ettore Di Giacinto 2026-04-01 11:50:17 +02:00 committed by GitHub
parent 88d5f8ffc3
commit e1cb817483
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/llama-memory-hybrid-iswa.cpp
View File

@ -73,9 +73,9 @@ llama_memory_context_ptr llama_memory_hybrid_iswa::init_batch(llama_batch_allocr
// if all tokens are output, split by sequence
ubatch = balloc.split_seq(n_ubatch);
} else {
// TODO: non-sequential equal split can be done if using unified KV cache
// for simplicity, we always use sequential equal split for now
ubatch = balloc.split_equal(n_ubatch, true);
// Use non-sequential split when KV cache is unified (needed for hellaswag/winogrande/multiple-choice)
const bool unified = (mem_attn->get_base()->get_n_stream() == 1);
ubatch = balloc.split_equal(n_ubatch, !unified);
}
if (ubatch.n_tokens == 0) {

6
src/llama-memory-hybrid.cpp
View File

@ -73,9 +73,9 @@ llama_memory_context_ptr llama_memory_hybrid::init_batch(llama_batch_allocr & ba
// if all tokens are output, split by sequence
ubatch = balloc.split_seq(n_ubatch);
} else {
// TODO: non-sequential equal split can be done if using unified KV cache
// for simplicity, we always use sequential equal split for now
ubatch = balloc.split_equal(n_ubatch, true);
// Use non-sequential split when KV cache is unified (needed for hellaswag/winogrande/multiple-choice)
const bool unified = (mem_attn->get_n_stream() == 1);
ubatch = balloc.split_equal(n_ubatch, !unified);
}
if (ubatch.n_tokens == 0) {