Commits · 799789afedcf03e3fc8ed2bbd12be77aac963256 · change / sglang

"examples/mxnet/sampling/vscode:/vscode.git/clone" did not exist on "08d4900fd10d8a24af5624ea020b1a52bc89409f"

30 Apr, 2025 6 commits
- Bump Flashinfer to 0.2.5 (#5870) · 799789af
  Baizhou Zhang authored Apr 29, 2025
```
Co-authored-by: Yuhao Chen <yxckeis8@gmail.com>
```
  799789af
- [PD] Fix Assertion failed: /DeepEP/csrc/kernels/internode.cu:483, condition:... · cc4a80ca
  ybyang authored Apr 30, 2025
```
[PD] Fix Assertion failed: /DeepEP/csrc/kernels/internode.cu:483, condition: ibgda_get_state()->num_rc_per_pe >= num_channels #134 (#5830)
```
  cc4a80ca
- Fix check_env script (#5901) · 3c8a5231
  lambert0312 authored Apr 30, 2025
  
  3c8a5231
- [Bugfix]: fix missing queue_time_start for requests from grammar_queue (#5696) · 28b26dbf
  Chang Su authored Apr 29, 2025
  
  28b26dbf
- feat: support pythonic tool call and index in tool call streaming (#5725) · 2b06484b
  Chang Su authored Apr 29, 2025
  
  2b06484b
- [fix] relax mem_fraction_static for h200 (#5893) · e4b6133b
  JieXin Liang authored Apr 30, 2025
```
Co-authored-by: alcanerian <alcanerian@gmail.com>
```
  e4b6133b
29 Apr, 2025 9 commits
- Auto set draft model path for MTP (#5793) · dd408ee4
  Ke Bao authored Apr 30, 2025
  
  dd408ee4
- Add A800 fused moe config for qwen3 30b (#5880) · 91dda4cd
  lambert0312 authored Apr 29, 2025
  
  91dda4cd
- [Fix] Fix a bug for flashmla to run R1 model (#5875) · 8e5a6d34
  pengcuo authored Apr 29, 2025
```
Co-authored-by: pengcuo <dgpengcuo@gmail.com>
```
  8e5a6d34
- Add qwen3 30b fused moe config (#5859) · 8465f035
  XinyuanTong authored Apr 29, 2025
  
  8465f035
- Feat: support cuda graph for LoRA (#4115) · 8c0cfca8
  Qiaolin Yu authored Apr 29, 2025
```
Co-authored-by: Beichen Ma <mabeichen12@gmail.com>
```
  8c0cfca8
- [Feature] support auto chat template (#4949) · 2c3ea294
  woodx authored Apr 29, 2025
  
  2c3ea294
- Cutlass MLA decode - fix dtype error (#5868) · 8d463fe3
  Trevor Morris authored Apr 28, 2025
  
  8d463fe3
- [CI] tune the test order to warmup the server (#5860) · 26fc32d1
  Lianmin Zheng authored Apr 28, 2025
  
  26fc32d1
- simplify fused_moe config logging (#5801) · 1cc32603
  Xiaoyu Zhang authored Apr 29, 2025
  
  1cc32603
28 Apr, 2025 20 commits
- Support `max_completion_tokens` for OpenAIChatCompletions (#5857) · 05ee2192
  Chang Su authored Apr 28, 2025
  
  05ee2192
- chore: bump v0.4.6.post1 (#5845) · dcae1fb2
  Yineng Zhang authored Apr 28, 2025
  
  dcae1fb2
- add fused moe config for qwen3moe fp8/bf16 (#5849) · a0251a3f
  Yi Zhang authored Apr 29, 2025
  
  a0251a3f
- feat: update is_fa3_default_architecture (#5854) · 663037a7
  Yineng Zhang authored Apr 28, 2025
  
  663037a7
- [Fix] Missing bootstrap_port field (#5823) · f4a9f60c
  XTY authored Apr 29, 2025
  
  f4a9f60c
- ROCm: update AITER (#5816) · d364b9b0
  HAI authored Apr 28, 2025
  
  d364b9b0
- [CI] test chunked prefill more (#5798) · 849c83a0
  Lianmin Zheng authored Apr 28, 2025
  
  849c83a0
- feat: Add fused moe triton config for qwen3-30b-fp8 moe on h20 (#5850) · d73ddeb1
  JiLi authored Apr 29, 2025
  
  d73ddeb1
- [config] qwen3moe_tune_h20 fp8 tp4 (#5846) · 74cb12a8
  ybyang authored Apr 29, 2025
  
  74cb12a8
- [PD] support pd fake transfer for warmup (#5726) · c6c62640
  ybyang authored Apr 29, 2025
  
  c6c62640
- feat: Add fused moe triton config for qwen3bf16 moe on h20 (#5839) · 92ab0a20
  yhyang201 authored Apr 29, 2025
  
  92ab0a20
- feat: Add fused moe triton config for qwen3 moe on h100 (#5833) · 0045f4b2
  XinyuanTong authored Apr 28, 2025
  
  0045f4b2
- fix: fix the error where the content is None when reasoning and tool … (#5838) · 8601300b
  mlmz authored Apr 28, 2025
  
  8601300b
- Feat: add support for thinking mode via chat_template_kwargs.enable_t… (#5551) · 6fa6f38e
  mlmz authored Apr 28, 2025
```
Co-authored-by: shuaills <shishuaiuoe@gmail.com>
Co-authored-by: Chayenne <zhaochen20@outlook.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
Co-authored-by: Yineng Zhang <me@zhyncs.com>
```
  6fa6f38e
- Revert "Tiny refactor DefaultModelLoader.Source" (#5825) · 693723d1
  Lianmin Zheng authored Apr 28, 2025
  
  693723d1
- Tiny refactor DefaultModelLoader.Source (#5482) · 644ed409
  fzyzcjy authored Apr 28, 2025
  
  644ed409
- Turn on overlap scheduler for multimodal models (#5771) · 3029889c
  Lianmin Zheng authored Apr 27, 2025
  
  3029889c
- chore: upgrade sgl-kernel 0.1.0 (#5690) · 41ac0c6d
  Yineng Zhang authored Apr 27, 2025
  
  41ac0c6d
- Add Cutlass MLA attention backend (#5390) · 84810da4
  Trevor Morris authored Apr 27, 2025
  
  84810da4
- Improve overlap scheduling (#5788) · 40d9b8ac
  Liangsheng Yin authored Apr 28, 2025
  
  40d9b8ac
27 Apr, 2025 5 commits
- [CI] Improve github summary & enable fa3 for more models (#5796) · daed453e
  Lianmin Zheng authored Apr 27, 2025
  
  daed453e
- Release v0.4.6 (#5795) · 84022c0e
  Baizhou Zhang authored Apr 27, 2025
  
  84022c0e
- [CI] Fix test case (#5790) · a38f6932
  Lianmin Zheng authored Apr 27, 2025
  
  a38f6932
- [PD]Reduce kv transfer threads (#5791) · beb65c74
  Liangsheng Yin authored Apr 27, 2025
  
  beb65c74
- [CI] Fix ci tests (#5769) · 621e96bf
  Lianmin Zheng authored Apr 27, 2025
  
  621e96bf