ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-28 18:32:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

Set mla=3 by default (#943)

Browse Source 
so more recent users that haven't followed the history of FlashMLA
evolution and hence don't know about the MLA options get the best setting
without having to add -mla 3 on the command line.

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2025-11-12 11:00:58 +02:00
committed by GitHub
parent 9ecfee6c03
commit 8a8de91a42
2 changed files with 12 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
common/common.h
View File

@@ -247,7 +247,7 @@ struct gpt_params {
bool simple_io = false; // improves compatibility with subprocesses and limited consoles
bool cont_batching = true; // insert new sequences for decoding on-the-fly
bool flash_attn = true; // flash attention
int mla_attn = 0; // MLA 0: standard attention, 1: MLA with K and transposed V cache, 2: MLA with just K cache
int mla_attn = 3; // MLA 0: standard, 1: MLA with K and V^T cache, 2: MLA with just K cache, 3: the best of both worlds
int attn_max_batch = 0; // Max batch size to use when computing attention (only applicable if flash_attn = false)
bool fused_moe_up_gate = true; // fused up*unary(gate) op for MoE models
bool fused_up_gate = true; // fused up*unary(gate) op