From 88c02fa108b56389a3fb26d5eda52f2dddc175c5 Mon Sep 17 00:00:00 2001 From: Kawrakow Date: Thu, 13 Nov 2025 09:52:06 +0200 Subject: [PATCH] Set default MLA to 3 also in llama-bench (#949) Co-authored-by: Iwan Kawrakow --- examples/llama-bench/llama-bench.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/llama-bench/llama-bench.cpp b/examples/llama-bench/llama-bench.cpp index 52b1dbbd..77327603 100644 --- a/examples/llama-bench/llama-bench.cpp +++ b/examples/llama-bench/llama-bench.cpp @@ -289,7 +289,7 @@ static const cmd_params cmd_params_defaults = { /* main_gpu */ {0}, /* no_kv_offload */ {false}, /* flash_attn */ {true}, - /* mla_attn */ {0}, + /* mla_attn */ {3}, /* attn_max_batch */ {0}, /* ser */ {{-1,0.0f}}, /* tensor_split */ {std::vector(llama_max_devices(), 0.0f)},