EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Add moe topk softmax templated from vllm (#4302)

Browse Source
This commit is contained in:
Qingquan Song
2025-03-14 12:03:33 -07:00
committed by GitHub
parent 660305c38a
commit 61e4433caf
9 changed files with 716 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
sgl-kernel/setup.py
View File

@@ -157,6 +157,7 @@ sources = [
"csrc/gemm/per_token_quant_fp8.cu",
"csrc/gemm/per_tensor_quant_fp8.cu",
"csrc/moe/moe_align_kernel.cu",
"csrc/moe/moe_topk_softmax_kernels.cu",
"csrc/speculative/eagle_utils.cu",
"csrc/speculative/speculative_sampling.cu",
"csrc/torch_extension.cc",