Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ggml: CUDA: add head size 72 for flash-attn (#16962)

Browse Source
This commit is contained in:
theo77186
2025-11-03 14:29:11 +01:00
committed by GitHub
parent 070ff4d535
commit 622cd010ff
5 changed files with 44 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/template-instances/fattn-tile-instance-dkq72-dv72.cu
+5
View File
@@ -0,0 +1,5 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-tile.cuh"
DECL_FATTN_TILE_CASE(72, 72);
ggml/src/ggml-cuda/template-instances/generate_cu_files.py
+3 -1
View File
@@ -3,7 +3,7 @@
from glob import glob
import os
HEAD_SIZES_KQ = [40, 64, 80, 96, 112, 128, 256, 576]
HEAD_SIZES_KQ = [40, 64, 72, 80, 96, 112, 128, 256, 576]
TYPES_KV = ["GGML_TYPE_F16", "GGML_TYPE_Q4_0", "GGML_TYPE_Q4_1", "GGML_TYPE_Q5_0", "GGML_TYPE_Q5_1", "GGML_TYPE_Q8_0"]
@@ -81,6 +81,8 @@ for ncols in [8, 16, 32, 64]:
for head_size_kq in HEAD_SIZES_KQ:
if head_size_kq == 40:
continue
if head_size_kq == 72:
continue
if head_size_kq != 576 and ncols2 == 16:
continue
if head_size_kq == 576 and ncols2 != 16: