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

opencl: transposed gemm/gemv moe kernel with mxfp4,f32 (#16602)

Browse Source 
* opencl: transposed gemm/gemv moe kernel with mxfp4,f32

* add restore kernel for moe transpose

* fix trailing whitespaces

* resolve compilation warnings
This commit is contained in:
Shawn Gu
2025-10-17 17:55:32 -07:00
committed by GitHub
parent 66b0dbcb2d
commit 81387858f1
5 changed files with 567 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-opencl/CMakeLists.txt
+2
View File
@@ -91,6 +91,8 @@ set(GGML_OPENCL_KERNELS
mul_mv_id_q8_0_f32_flat
mul_mv_id_mxfp4_f32
mul_mv_id_mxfp4_f32_flat
gemm_moe_mxfp4_f32
gemv_moe_mxfp4_f32
mul_mm_f32_f32_l4_lm
mul_mm_f16_f32_l4_lm
mul_mm_q8_0_f32_l4_lm