Superminaren/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-07 20:47:30 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix: update mmq_use_routed_moe_ncols_picker to include NVIDIA + Volta support

Browse Source
This commit is contained in:
ravel7524
2026-09-01 13:04:01 +02:00
committed by Carl Philipp Klemm
parent a76a341dcf
commit ddd960669e
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/mmq.cuh
+2 -1
View File
@@ -1473,7 +1473,8 @@ static void launch_mul_mat_q(ggml_backend_cuda_context & ctx, const mmq_args & a
}
static bool mmq_use_routed_moe_ncols_picker(const int cc) {
return GGML_CUDA_CC_IS_CDNA(cc) ||
return (GGML_CUDA_CC_IS_NVIDIA(cc) && cc >= GGML_CUDA_CC_VOLTA) ||
GGML_CUDA_CC_IS_CDNA(cc) ||
GGML_CUDA_CC_IS_RDNA2(cc) ||
GGML_CUDA_CC_IS_RDNA3(cc) ||
GGML_CUDA_CC_IS_RDNA4(cc);