From 9882f21ab938e24f8108e68a142b62f95992ba8e Mon Sep 17 00:00:00 2001 From: ingyukoh Date: Sun, 11 Jan 2026 01:44:36 +0900 Subject: [PATCH] server: add missing rerank and chat presets (#10932) --- common/arg.cpp | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/common/arg.cpp b/common/arg.cpp index 72750a3cba..bd736a0eda 100644 --- a/common/arg.cpp +++ b/common/arg.cpp @@ -3505,6 +3505,30 @@ common_params_context common_params_parser_init(common_params & params, llama_ex } ).set_examples({LLAMA_EXAMPLE_TTS})); + add_opt(common_arg( + {"--rerank-bge-m3-default"}, + string_format("use BAAI/bge-reranker-v2-m3 (note: can download weights from the internet)"), + [](common_params & params) { + params.model.hf_repo = "BAAI/bge-reranker-v2-m3"; + params.model.hf_file = "bge-reranker-v2-m3-Q8_0.gguf"; + params.embedding = true; + params.port = 8015; + } + ).set_examples({LLAMA_EXAMPLE_SERVER})); + + add_opt(common_arg( + {"--chat-llama3.1-8b-default"}, + string_format("use Meta Llama 3.1 8B Instruct (note: can download weights from the internet)"), + [](common_params & params) { + params.model.hf_repo = "gaianet/Meta-Llama-3.1-8B-Instruct-GGUF"; + params.model.hf_file = "Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf"; + params.port = 8016; + params.n_ubatch = 512; + params.n_batch = 2048; + params.n_ctx = 8192; + } + ).set_examples({LLAMA_EXAMPLE_SERVER})); + add_opt(common_arg( {"--embd-gemma-default"}, string_format("use default EmbeddingGemma model (note: can download weights from the internet)"),