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

rebase h20 fused_moe config (#6966)

Browse Source
This commit is contained in:
Xiaoyu Zhang
2025-06-08 20:01:34 +08:00
committed by GitHub
parent 608668e143
commit fa3592cfeb
4 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
benchmark/kernels/fused_moe_triton/README.md
View File

@@ -42,7 +42,7 @@ python benchmark/kernels/fused_moe_triton/tuning_fused_moe_triton.py \
--tune
```
After tuning, a configuration file (e.g., `E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json`) will be generated in the current directory. You can move this file to `sglang/srt/layers/fused_moe_triton/configs/` to use it in `sglang`.
After tuning, a configuration file (e.g., `E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json`) will be generated in the current directory. You can move this file to `sglang/srt/layers/fused_moe_triton/configs/triton_version` dir to use it in `sglang`.
### Performance Comparison Tool

0
python/sglang/srt/layers/moe/fused_moe_triton/configs/README → python/sglang/srt/layers/moe/fused_moe_triton/configs/README.md
View File

0
python/sglang/srt/layers/moe/fused_moe_triton/configs/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
View File

0
python/sglang/srt/layers/moe/fused_moe_triton/configs/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
View File