Commits · cb3918a09127a1da8cc0976f86e7425285a1dca6 · change / sglang

07 Sep, 2025 1 commit

Optimize moe_sum_reduce_kernel (#9477) · cb3918a0

Yuan Luo authored Sep 07, 2025


Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
Co-authored-by: Xiaoyu Zhang <35585791+BBuf@users.noreply.github.com>

cb3918a0

03 Sep, 2025 1 commit
- [benchmark] add flashinfer_allreduce_fusion benchmark (#9937) · b1fb7e45
  Xiaoyu Zhang authored Sep 03, 2025
  
  b1fb7e45
30 Aug, 2025 1 commit
- [NVIDIA] [2/N] Optimize `silu_and_mul_scaled_fp4_grouped_quant` perf (#9556) · 5c34b4f1
  Kaixi Hou authored Aug 29, 2025
  
  5c34b4f1
27 Aug, 2025 1 commit
- Add A100 fused MoE kernel configs for Dpsk (#9677) · 8f7b1c31
  ehuaa authored Aug 27, 2025
  
  8f7b1c31
25 Aug, 2025 1 commit
- fix: resolve tuning fused moe issue (#9587) · f8b757bc
  Yineng Zhang authored Aug 25, 2025
  
  f8b757bc
20 Aug, 2025 2 commits

Revert "[feature] Rework Ascend NPU graph support" (#9385) · de2dd738
Even Zhou authored Aug 20, 2025

de2dd738

[feature] Rework Ascend NPU graph support (#9350) · 3680d6f8

Even Zhou authored Aug 20, 2025


Co-authored-by: ronnie_zheng <zl19940307@163.com>
Co-authored-by: yezhifeng (D) <y00897525@china.huawei.com>
Co-authored-by: anon189Ty <Stari_Falcon@outlook.com>
Co-authored-by: Maksim <makcum888e@mail.ru>
Co-authored-by: ssshinigami <44640852+ssshinigami@users.noreply.github.com>

3680d6f8

19 Aug, 2025 3 commits
- [CI] Fix lint issues (#9361) · 46fe8b8c
  Chang Su authored Aug 19, 2025
  
  46fe8b8c
- fix: enable multi-GPU Triton fused MoE tuning (#6295) · a3b810eb
  mpashkovskiy authored Aug 19, 2025
  
  a3b810eb
- Revert "[feature] Ascend NPU graph support (#8027)" (#9348) · f4fafacc
  Even Zhou authored Aug 20, 2025
  
  f4fafacc
18 Aug, 2025 1 commit
- Fix triton_fused_moe unit test and benchmark (#9276) · 968e1818
  Yuan Luo authored Aug 18, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  968e1818
17 Aug, 2025 1 commit

[feature] Ascend NPU graph support (#8027) · 94371dbb

VDV1985 authored Aug 17, 2025


Co-authored-by: ronnie_zheng <zl19940307@163.com>
Co-authored-by: yezhifeng (D) <y00897525@china.huawei.com>
Co-authored-by: anon189Ty <Stari_Falcon@outlook.com>
Co-authored-by: Maksim <makcum888e@mail.ru>
Co-authored-by: ssshinigami <44640852+ssshinigami@users.noreply.github.com>

94371dbb

15 Aug, 2025 1 commit
- [6/N] MoE Refactor: Cleanup MoE-related configs (#8849) · 29589512
  Cheng Wan authored Aug 14, 2025
  
  29589512
06 Aug, 2025 1 commit
- Optimize triton swa kernel by skipping computation (#8860) · 0475448e
  Ke Bao authored Aug 06, 2025
  
  0475448e
28 Jul, 2025 2 commits
- feat: support glm4 tuning (#8473) · 1466c1b8
  Yineng Zhang authored Jul 28, 2025
  
  1466c1b8
- GLM-4.5 Model Support (#8224) · 6d6a8bc2
  Yuxuan Zhang authored Jul 28, 2025
```
Co-authored-by: Lifu Huang <lifu.hlf@gmail.com>
Co-authored-by: Binyao Jiang <byjiang1996@gmail.com>
Co-authored-by: Stefan He <hebiaobuaa@gmail.com>
```
  6d6a8bc2
20 Jul, 2025 1 commit
- Fix tuning_fused_moe_triton.py (#8175) · abda2542
  Cheng Wan authored Jul 19, 2025
  
  abda2542
07 Jul, 2025 1 commit
- Integrate triton moe kernel (#7689) · 253454de
  Yuan Luo authored Jul 07, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  253454de
16 Jun, 2025 1 commit
- refine fused_moe benchmark (#7221) · 0ae1e9a7
  Xiaoyu Zhang authored Jun 16, 2025
  
  0ae1e9a7
12 Jun, 2025 1 commit
- Fix positional argument (#7093) · ef326774
  Quanfeng Li authored Jun 12, 2025
  
  ef326774
08 Jun, 2025 3 commits
- Fuse routed scaling factor in deepseek (#6970) · 3712abfa
  Xiaoyu Zhang authored Jun 09, 2025
  
  3712abfa
- rebase h20 fused_moe config (#6966) · fa3592cf
  Xiaoyu Zhang authored Jun 08, 2025
  
  fa3592cf
- Revert "Fuse routed scaling factor in topk_reduce kernel (#6220)" (#6968) · 1fb76ebb
  Yineng Zhang authored Jun 07, 2025
  
  1fb76ebb
07 Jun, 2025 1 commit
- Fuse routed scaling factor in topk_reduce kernel (#6220) · 515ef4fa
  Xiaoyu Zhang authored Jun 08, 2025
  
  515ef4fa
05 Jun, 2025 1 commit
- support 1 shot allreduce in 1-node and 2-node using mscclpp (#6277) · 8e3797be
  zyksir authored Jun 05, 2025
  
  8e3797be
04 Jun, 2025 2 commits
- Set `num_fused_shared_experts` as `num_shared_experts` when shared_experts... · 81964328
  Cheng Wan authored Jun 04, 2025
```
Set `num_fused_shared_experts` as `num_shared_experts` when shared_experts fusion is not disabled (#6736)
```
  81964328
- [Refactor] Rename `n_share_experts_fusion` as `num_fused_shared_experts` (#6735) · 8a548052
  Cheng Wan authored Jun 03, 2025
  
  8a548052
29 May, 2025 3 commits
- [test] add ut and bm for get_last_loc (#6746) · d9d35def
  JieXin Liang authored May 30, 2025
  
  d9d35def
- Support tuning DeepEP configs (#6742) · 6df81e8a
  fzyzcjy authored May 29, 2025
  
  6df81e8a
- refactor apply_w8a8_block_fp8_linear in fp (#6545) · 485a023b
  ChangyiYang authored May 29, 2025
  
  485a023b
28 May, 2025 2 commits
- fix log_info_on_rank0 error when run benchmark (#6260) · 07610353
  Xiaoyu Zhang authored May 28, 2025
  
  07610353
- Refine pre_reorder_triton_kernel slightly to improve performance (#6627) · c087ddd6
  Yuan Luo authored May 28, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  c087ddd6
12 May, 2025 3 commits
- Support tuning moe for llama 4 model (#6042) · ef8ec07b
  fzyzcjy authored May 13, 2025
  
  ef8ec07b
- 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
- Replace time.time() to time.perf_counter() for benchmarking. (#6178) · 6e2da515
  Lifu Huang authored May 11, 2025
```
Signed-off-by: Lifu Huang <lifu.hlf@gmail.com>
```
  6e2da515
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