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

opencl: add q4_1 MoE for Adreno (#22856)

Browse Source 
* Q4_1 MoE CLC pass sanity check

* remove unnecessary code

* opencl: remove unnecessary asserts and reformat

* opencl: fix supports_op for q4_1 moe

* q4_1 moe is supported by Adreno with certain shapes

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>
This commit is contained in:
Shawn Gu
2026-05-11 11:57:26 -07:00
committed by GitHub
co-authored by Li He
parent 8e1f9d0834
commit 1ec7ba0c14
5 changed files with 798 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-opencl/CMakeLists.txt
+2
View File
@@ -104,6 +104,8 @@ set(GGML_OPENCL_KERNELS
mul_mv_id_mxfp4_f32_flat
gemm_moe_q4_0_f32_ns
gemv_moe_q4_0_f32_ns
gemm_moe_q4_1_f32_ns
gemv_moe_q4_1_f32_ns
gemm_moe_mxfp4_f32
gemv_moe_mxfp4_f32
gemm_moe_mxfp4_f32_ns