Commits · e9a47f4cb58a5a2fedd7843211684b8e4db3c0c5 · change / sglang

29 Apr, 2025 1 commit
- simplify fused_moe config logging (#5801) · 1cc32603
  Xiaoyu Zhang authored Apr 29, 2025
  
  1cc32603
28 Apr, 2025 3 commits
- add fused moe config for qwen3moe fp8/bf16 (#5849) · a0251a3f
  Yi Zhang authored Apr 29, 2025
  
  a0251a3f
- fused moe triton tuning script support qwen3 (#5842) · e132cba2
  Xiaoyu Zhang authored Apr 29, 2025
  
  e132cba2
- feat: Add fused moe triton config for qwen3 moe on h100 (#5833) · 0045f4b2
  XinyuanTong authored Apr 28, 2025
  
  0045f4b2
20 Apr, 2025 1 commit
- Minor update for ROCm variable style (#5562) · c555d794
  Zhaoyi Li authored Apr 20, 2025
  
  c555d794
15 Apr, 2025 1 commit
- Add A800 shared experts fused MoE kernel tuning configs for DeepSeek V3/R1 (#5368) · 61e7c4dd
  lambert0312 authored Apr 15, 2025
  
  61e7c4dd
12 Apr, 2025 1 commit
- refine fused_moe tuning docs (#5294) · 3e4794aa
  Xiaoyu Zhang authored Apr 13, 2025
  
  3e4794aa
04 Apr, 2025 1 commit
- Add DeepSeek V3/R1 shared experts fusion (#4918) · 924ca7c9
  Xiaoyu Zhang authored Apr 04, 2025
  
  924ca7c9
28 Mar, 2025 1 commit
- [CI] Remove unused imports with Ruff to pre-commit config, only to... · b149b393
  Brayden Zhong authored Mar 27, 2025
```
[CI] Remove unused imports with Ruff to pre-commit config, only to benchmarks/docs/examples folder (#3969)
```
  b149b393
25 Mar, 2025 1 commit
- [Benchmark] tilelang vs deepgemm vs w8a8_block_fp8_matmul (#4735) · 14269198
  Chunan Zeng authored Mar 24, 2025
  
  14269198
24 Mar, 2025 1 commit
- rename benchmark_deepgemm_fp8_group_gemm.py (#4605) · 3980ff1b
  Tongbao Zhang authored Mar 24, 2025
  
  3980ff1b
22 Mar, 2025 1 commit
- Correcting default configuration when benchmarking fused_moe (#4665) · 38f25e87
  penguin_wwy authored Mar 22, 2025
  
  38f25e87
16 Mar, 2025 1 commit
- [Fix] use `torch.cat` instead of `torch.concat` to prevent entering the... · 1a3fa75f
  JieXin Liang authored Mar 16, 2025
```
[Fix] use `torch.cat` instead of `torch.concat` to prevent entering the `Autograd` backends. (#4466)
```
  1a3fa75f
12 Mar, 2025 1 commit
- refine sgl_moe_align_block_size_benchmark (#4327) · 7130a7ce
  Xiaoyu Zhang authored Mar 12, 2025
  
  7130a7ce
11 Mar, 2025 1 commit
- Add A100 tuning configs for DeepSeek R1/V3 channel-wise INT8 (#4287) · 6a02b32d
  yych0745 authored Mar 11, 2025
```
Co-authored-by: HandH1998 <1335248067@qq.com>
```
  6a02b32d
03 Mar, 2025 4 commits
- Improve code styles (#4021) · 66301e12
  Lianmin Zheng authored Mar 03, 2025
  
  66301e12
- Support penalty in overlap mode; return logprob with chunked prefill; improve... · ac238727
  Lianmin Zheng authored Mar 03, 2025
```
Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: dhou-xai <dhou@x.ai>
Co-authored-by: Hanming Lu <hanming_lu@berkeley.edu>
```
  ac238727
- Optimize Triton Kernel of Group GEMM in DeepGEMM Benchmark (#4014) · 0194948f
  Stefan He authored Mar 02, 2025
  
  0194948f
- Add Benchmark for DeepGEMM Group GEMM (#3993) · b7e274f2
  Stefan He authored Mar 02, 2025
  
  b7e274f2
02 Mar, 2025 2 commits
- fix typo in deep gemm benchmarking(#3991) · 50f28f65
  Xiaoyu Zhang authored Mar 02, 2025
  
  50f28f65
- add deepgemm and sglang fp8 block-wise gemm benchmark (#3893) · 90a55e25
  Xiaoyu Zhang authored Mar 02, 2025
  
  90a55e25
01 Mar, 2025 1 commit
- Revert "[MOE] enable efficient moe_alignment multi-blocks execution (3x~6x)" (#3982) · 18bb216c
  Chayenne authored Feb 28, 2025
  
  18bb216c
28 Feb, 2025 1 commit
- [MOE] enable efficient moe_alignment multi-blocks execution (3x~6x) (#3613) · 1c96fa86
  yiakwy-xpu-ml-framework-team authored Feb 28, 2025
  
  1c96fa86
27 Feb, 2025 1 commit
- Tuning Script for Feature DeepSeek V3/R1 INT8 Quantization (block-wise) (#3922) · b0df5d24
  laixin authored Feb 27, 2025
```
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
```
  b0df5d24
18 Feb, 2025 1 commit
- [ROCm] Optimal MOE Tuning for AMD Radeon Graphics (#3567) · ddf39d3f
  yigex authored Feb 18, 2025
  
  ddf39d3f
17 Feb, 2025 1 commit
- support multi-gpu block-gemm tuning (#3639) · c38f3aed
  Xiaoyu Zhang authored Feb 18, 2025
  
  c38f3aed
11 Feb, 2025 1 commit
- [ROCm] Add ROCm tuning config to block gemm and Re-tune for AMD Radeon Graphics (#3418) · fdf04a14
  yigex authored Feb 11, 2025
```
Co-authored-by: Bruce Xue <yigex@xilinx.com>
Co-authored-by: HAI <hixiao@gmail.com>
```
  fdf04a14
10 Feb, 2025 1 commit
- refine some typo (#3473) · 2f47d710
  Xiaoyu Zhang authored Feb 10, 2025
  
  2f47d710
08 Feb, 2025 3 commits
- fix EAGLE 2 non greedy case (#3407) · fad315cb
  Yineng Zhang authored Feb 09, 2025
```
Co-authored-by: Ying Sheng <sqy1415@gmail.com>
```
  fad315cb
- Update fused_moe's benchmark (#3346) · 849f58d6
  GaoYuYang authored Feb 08, 2025
  
  849f58d6
- [BUG] fix moe benchmark when bs*seq is small (#3382) · 64480df4
  yiakwy-xpu-ml-framework-team authored Feb 08, 2025
  
  64480df4
06 Feb, 2025 2 commits
- optimize moe_align_kernel cuda (#3347) · cdae77b0
  Xiaoyu Zhang authored Feb 07, 2025
  
  cdae77b0
- clean moe align block kernel code and add acc test (#3332) · ad349985
  Xiaoyu Zhang authored Feb 06, 2025
  
  ad349985
31 Jan, 2025 1 commit
- add tuning block wise fp8 (#3242) · 7b020cca
  Yineng Zhang authored Feb 01, 2025
```
Co-authored-by: HandH1998 <007aabbcc411@gmail.com>
```
  7b020cca
27 Jan, 2025 1 commit
- add dsv3 mi300 triton config for block scale (#3146) · 351a72d4
  yigex authored Jan 27, 2025
  
  351a72d4
26 Jan, 2025 1 commit
- Use torch.compile for scaling penalty (#3133) · 27acf63b
  Lianmin Zheng authored Jan 25, 2025
  
  27acf63b
23 Jan, 2025 1 commit
- support lightning_attention_decode in sgl-kernel for MiniMax-Text-01 (#3030) · ac2dc35d
  Xiaoyu Zhang authored Jan 23, 2025
  
  ac2dc35d
20 Jan, 2025 1 commit
- fix moe align blocks benchmark (#3003) · 10bfce71
  yiakwy-xpu-ml-framework-team authored Jan 20, 2025
  
  10bfce71
19 Jan, 2025 1 commit
- fix file name spelling mistake and useless variable in minmax-text-01-lightning_attention (#2971) · 83452dbb
  Xiaoyu Zhang authored Jan 19, 2025
  
  83452dbb
18 Jan, 2025 1 commit
- optimize MiniMax-Text-01 lightning_attn_decode triton (#2966) · c2f212d6
  Xiaoyu Zhang authored Jan 18, 2025
  
  c2f212d6