apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

【Inference Optimize】Support automatic generation of marlin kernel (#3149)

Browse Source 
* Support automatic generation of marlin kernel
This commit is contained in:
AIbin
2025-08-01 22:43:18 +08:00
committed by GitHub
parent b71cbb466d
commit 22fe695f1c
7 changed files with 125 additions and 376 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
custom_ops/setup_ops.py
View File

@@ -409,6 +409,7 @@ elif paddle.is_compiled_with_cuda():
sources += find_end_files("gpu_ops/speculate_decoding", ".cc")
nvcc_compile_args += ["-DENABLE_BF16"]
# moe
os.system("python gpu_ops/moe/moe_wna16_marlin_utils/generate_kernels.py")
sources += find_end_files("gpu_ops/cutlass_kernels/moe_gemm/", ".cu")
sources += find_end_files("gpu_ops/cutlass_kernels/w4a8_moe/", ".cu")
sources += find_end_files("gpu_ops/moe/", ".cu")