Commits · 2998c4bdf4fea00081978b320b60f4b0ce905297 · change / sglang

03 Jul, 2025 2 commits
- [optimize] fuse renormalize into moe_topk_softmax (#7744) · 2998c4bd
  Yi Zhang authored Jul 04, 2025
```
Co-authored-by: ispobock <ispobaoke@gmail.com>
```
  2998c4bd
- Add CUTLASS FP8 Blockscale MoE kernel for Hopper architecture (#7278) · 2c4feaf3
  ayrnb authored Jul 03, 2025
```
Co-authored-by: HydraQYH <QYH820@Outlook.com>
Co-authored-by: TianQiLin666666 <1834987979@qq.com>
```
  2c4feaf3
30 Jun, 2025 1 commit
- Add dsv3 router gemm kernel (#7627) · 7248272c
  Baizhou Zhang authored Jun 29, 2025
  
  7248272c
29 Jun, 2025 1 commit
- Add dsv3 fused a gemm to sgl-kernel (#7630) · 04b35190
  Ke Bao authored Jun 29, 2025
  
  04b35190
25 Jun, 2025 1 commit
- Fuse sorted_token_ids padding to moe_align_block_size kernel (#7437) · 57ab7769
  Ke Bao authored Jun 25, 2025
  
  57ab7769
23 Jun, 2025 1 commit

feat: integrate deepgemm into EPMoE (#6821) · 506c4928

xutizhou authored Jun 23, 2025


Co-authored-by: tianqilin.99 <tianqilin.99@bytedance.com>
Co-authored-by: TianQiLin666666 <1834987979@qq.com>
Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>

506c4928

14 Jun, 2025 1 commit
- [fix] fix cutlass_mla_backend with cuda_graph and add sm_scale for sgl-kernel cutlass_mla (#7184) · ab1a4fa5
  JieXin Liang authored Jun 15, 2025
  
  ab1a4fa5
13 Jun, 2025 1 commit
- Remove 200us slow concat kernel (part 1: kernel) (#7145) · aa46ed34
  fzyzcjy authored Jun 13, 2025
  
  aa46ed34
12 Jun, 2025 1 commit
- [sgl-kernel] Add cuda kernel for moe_ep_silu_and_mul (#6919) · 84727a51
  Yuan Luo authored Jun 12, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  84727a51
09 Jun, 2025 1 commit
- [perf][sgl-kernel] extend cutlass_mla_decode to support num_head < 128 (#6929) · 18efb5e8
  JieXin Liang authored Jun 09, 2025
  
  18efb5e8
05 Jun, 2025 1 commit
- [EP] Add cuda kernel for moe_ep_post_reorder (#6837) · 43baba64
  Yuan Luo authored Jun 05, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  43baba64
04 Jun, 2025 1 commit
- fix ep_moe_reorder kernel bugs (#6858) · bd75690f
  Xiaoyu Zhang authored Jun 04, 2025
```
Co-authored-by: JieXin Liang <Alcanderian@users.noreply.github.com>
```
  bd75690f
02 Jun, 2025 1 commit
- [EP] Add cuda kernel for moe_ep_pre_reorder (#6699) · 55444ed6
  Yuan Luo authored Jun 02, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  55444ed6
29 May, 2025 1 commit
- refactor apply_w8a8_block_fp8_linear in fp (#6545) · 485a023b
  ChangyiYang authored May 29, 2025
  
  485a023b
22 May, 2025 1 commit

[1/2] Support Qserve (#6457) · 4d643f6c

HandH1998 authored May 22, 2025


Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>

4d643f6c

12 May, 2025 2 commits
- Revert "fix some typos" (#6244) · e8e18dcd
  Lianmin Zheng authored May 12, 2025
  
  e8e18dcd
- fix some typos (#6209) · d738ab52
  applesaucethebun authored May 12, 2025
```
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
```
  d738ab52
11 May, 2025 1 commit
- Add typo checker in pre-commit (#6179) · 2ce87935
  applesaucethebun authored May 11, 2025
```
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
```
  2ce87935
12 Apr, 2025 1 commit
- update variable naming and comments for rocm (#5299) · 3c9740d2
  Zhaoyi Li authored Apr 12, 2025
  
  3c9740d2
11 Apr, 2025 1 commit
- fix: remove cublas_grouped_gemm (#5307) · 136b8e6a
  Yineng Zhang authored Apr 11, 2025
  
  136b8e6a
10 Apr, 2025 1 commit
- reduce moe_align_block_size_kernel small batch mode overhead (#5086) · f730362e
  Xiaoyu Zhang authored Apr 10, 2025
  
  f730362e
09 Apr, 2025 1 commit
- sgl-kernel use cutlass latest version for fp8 blockwise gemm (#5207) · ebf495f0
  Yi Zhang authored Apr 10, 2025
  
  ebf495f0
03 Apr, 2025 1 commit
- [sgl-kernel] per token group quant support COLUMN MAJOR (#4817) · 2c8fd993
  Xiaoyu Zhang authored Apr 03, 2025
  
  2c8fd993
29 Mar, 2025 1 commit
- Add deepseek style fused moe group gate selection kernel (#4530) · 45dcfc2e
  Qingquan Song authored Mar 29, 2025
  
  45dcfc2e
24 Mar, 2025 1 commit
- [Quant Kernel] refactored per token group quant fp8 to support int8 up-to 2x faster (#4396) · 65c24c28
  Chunan Zeng authored Mar 23, 2025
  
  65c24c28
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
14 Mar, 2025 1 commit
- Add moe topk softmax templated from vllm (#4302) · 61e4433c
  Qingquan Song authored Mar 14, 2025
  
  61e4433c
13 Mar, 2025 3 commits
- fix accuracy issue (#4376) · 2937387a
  Yineng Zhang authored Mar 13, 2025
  
  2937387a
- Fix per token fp8 quant precision (#4362) · 4068e012
  Qingquan Song authored Mar 12, 2025
  
  4068e012
- feat: support ep size < 32 for sgl kernel (#4348) · 817d4370
  Shi Shuai authored Mar 13, 2025
  
  817d4370
12 Mar, 2025 2 commits
- Add awq dequantize kernel to sgl with 1x to 3x speedup (#4104) · 07f94463
  Rex authored Mar 12, 2025
  
  07f94463
- refine sgl_moe_align_block_size_benchmark (#4327) · 7130a7ce
  Xiaoyu Zhang authored Mar 12, 2025
  
  7130a7ce
07 Mar, 2025 3 commits
- [Refactor] Reducing code duplication across FP8 CUDA quantization kernels (#4163) · 95085d65
  Stefan He authored Mar 06, 2025
  
  95085d65
- Add sgl_per_token_quant_fp8 (#4089) · 63ee26d1
  Stefan He authored Mar 06, 2025
  
  63ee26d1
- [quant kernel] sgl-kernel support per_tensor_quant fp8 (#3786) · ad55f171
  Xiaoyu Zhang authored Mar 07, 2025
  
  ad55f171
19 Feb, 2025 1 commit
- use warp shuffle style reduce and flashinfer vectorize (#3628) · 55a7ec38
  Xiaoyu Zhang authored Feb 19, 2025
  
  55a7ec38
18 Feb, 2025 1 commit
- [Feature] Apply Cublas Grouped Gemm kernel (#3629) · 67fc595b
  Baizhou Zhang authored Feb 17, 2025
  
  67fc595b
12 Feb, 2025 1 commit
- support blockwise fp8 matmul kernel (#3267) · 640363ad
  yizhang2077 authored Feb 13, 2025
  
  640363ad
11 Feb, 2025 1 commit
- optimize per token group quant fp8 (#3490) · bb418ced
  Xiaoyu Zhang authored Feb 11, 2025
  
  bb418ced
28 Jan, 2025 1 commit
- clean up useless file (#3192) · 81262c7b
  Xiaoyu Zhang authored Jan 28, 2025
  
  81262c7b