Commits · 2810338401d0d5abe8ea89be45d69728e6d22602 · change / sglang

"vscode:/vscode.git/clone" did not exist on "f949ad5794c17d34fe1fb90d34143764e42cc86a"

28 Jul, 2025 4 commits
- [feat] Support different attention backends for prefill and decode (#6338) · 28103384
  Qiaolin Yu authored Jul 27, 2025
```
Co-authored-by: tianqilin.99 <tianqilin.99@bytedance.com>
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
```
  28103384
- [router] improve router logs and request id header (#8415) · fe6a445d
  Simo Lin authored Jul 27, 2025
  
  fe6a445d
- bugfix: Fix XGrammar backend to use model's EOS tokens for constrained generation (#8422) · dd487e55
  Chang Su authored Jul 27, 2025
  
  dd487e55
- Fix docker buildx push error (#8425) · bb81daef
  kyleliang-nv authored Jul 27, 2025
  
  bb81daef
27 Jul, 2025 28 commits
- Fix test_openai_server (#8419) · 58dd95fb
  Chang Su authored Jul 27, 2025
  
  58dd95fb
- bugfix: Fix multiple finish_reason chunks and tool_calls finish reason check (#8417) · b47eda33
  Chang Su authored Jul 27, 2025
  
  b47eda33
- Fix: Improve test_openai_function_calling unit test and fix... · e983d666
  Binyao Jiang authored Jul 27, 2025
```
Fix: Improve test_openai_function_calling unit test and fix reasoning_parser.py think_start_token logic (#8316)
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  e983d666
- Support ue8m0 for triton quant kernel (#7603) · b58c3c28
  fzyzcjy authored Jul 28, 2025
  
  b58c3c28
- Support overlapped lora updates (#8213) · df906455
  Lifu Huang authored Jul 27, 2025
  
  df906455
- Change to use native arm runner (#8414) · 95217a9b
  kyleliang-nv authored Jul 27, 2025
  
  95217a9b
- [Bugfix] Prevent PD server crash from invalid grammar (#8062) · 22e00eeb
  Shangming Cai authored Jul 28, 2025
```
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
```
  22e00eeb
- Support triton kernels v3.4.0 for fused_moe (#8258) · b3eac168
  Yuan Luo authored Jul 27, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
Co-authored-by: Cheng Wan <cwan@x.ai>
Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>
```
  b3eac168
- chore: upgrade flashinfer v0.2.9rc2 (#8406) · 10ee8955
  Yineng Zhang authored Jul 27, 2025
  
  10ee8955
- chore: update CODEOWNERS (#8407) · bf3352c5
  Yineng Zhang authored Jul 27, 2025
  
  bf3352c5
- [hotfix] fix merge conflicts in FlashInferEPMoE (#8405) · 4d921f2b
  Cheng Wan authored Jul 27, 2025
  
  4d921f2b
- Support precomputed_embeddings for Llama 4 (#8156) · 44d600cd
  Kevin Xiang Li authored Jul 27, 2025
```
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: Xiang (Kevin) Li <lik@nvidia.com>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
```
  44d600cd
- Use FlashInfer FP4 gemm. (#8241) · 5c9c275b
  Elfie Guo authored Jul 27, 2025
  
  5c9c275b
- [2/N] MoE Refactor: Unify weight loader and quant methods (#8397) · bf0f448f
  Cheng Wan authored Jul 27, 2025
  
  bf0f448f
- fix: fix the missing metrics on non-rank0 nodes (#7720) · 36d6f0ba
  Yingchun Lai authored Jul 27, 2025
  
  36d6f0ba
- Add A800 fused MoE kernel tuning configs for Qwen3-Coder-480B-A35B-Instruct (#8351) · 2a1936de
  Li Hui authored Jul 27, 2025
  
  2a1936de
- [router] add different policies for p node and d node (#8395) · 2ab97023
  Simo Lin authored Jul 27, 2025
  
  2ab97023
- fix: minor fix TransportProxyTensor under tp (#8382) · 0bcc195f
  Mick authored Jul 27, 2025
  
  0bcc195f
- Update Cutlass in sgl-kernel to v4.1 (#8392) · 91e3d154
  Baizhou Zhang authored Jul 27, 2025
  
  91e3d154
- [NVIDIA] Add Flashinfer MoE blockscale fp8 backend (#8036) · 85486b6f
  Kaixi Hou authored Jul 27, 2025
  
  85486b6f
- Fix bench script making input data on L2 cache (#7739) · e34cf6ad
  fzyzcjy authored Jul 27, 2025
  
  e34cf6ad
- Minor tool for comparison of benchmark results (#7974) · 62222bd2
  fzyzcjy authored Jul 27, 2025
  
  62222bd2
- Tool to dump and compare internal activation tensors (#7976) · ed0fdbf3
  fzyzcjy authored Jul 27, 2025
  
  ed0fdbf3
- Urgent Fix: intern-s1 chat-template matching (#8403) · b602f423
  Xinyuan Tong authored Jul 27, 2025
```
Signed-off-by: Xinyuan Tong <justinning0323@outlook.com>
```
  b602f423
- Add nvfp4 scaled mm benchmark. (#8401) · 426b7493
  Qi Yuhang authored Jul 27, 2025
  
  426b7493
- HiCache, check before terminate prefetching (#8372) · 528bd1ed
  Zhiqiang Xie authored Jul 26, 2025
  
  528bd1ed
- Add docker release flow for gb200 (#8394) · 62a6b7c7
  kyleliang-nv authored Jul 26, 2025
  
  62a6b7c7
- Remove slot usage in code to be backward-compatible with python 3.9 (#8396) · 76154631
  Lifu Huang authored Jul 26, 2025
  
  76154631
26 Jul, 2025 8 commits
- Add perf tests for LoRA (#8314) · 5c705b1d
  Lifu Huang authored Jul 26, 2025
  
  5c705b1d
- model: support intern-s1 (#8350) · b7094a5e
  RunningLeon authored Jul 27, 2025
```
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: zxy <zhou0493@e.ntu.edu.sg>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
```
  b7094a5e
- Tiny assert EPLB is used together with expert parallel (#8381) · da0c0260
  fzyzcjy authored Jul 26, 2025
  
  da0c0260
- vlm: optimize tensor transport (#6003) · 3212c2ad
  Mick authored Jul 26, 2025
```
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
```
  3212c2ad
- chore: improvements on mm_utils (#7737) · 53475674
  Mick authored Jul 26, 2025
  
  53475674
- Extract update_weights from RL Engine to SGLang to keep simplicity and fix torch reduce (#8267) · ce32bc2b
  Stefan He authored Jul 26, 2025
```
Co-authored-by: CuiBo 82354186+SuperCB@users.noreply.github.com
Co-authored-by: GeLee 865038696@qq.com
Co-authored-by: 杨睿 yangruipis@163.com
```
  ce32bc2b
- Save peak memory in logits processor (#8343) · e236d8fe
  Cheng Wan authored Jul 26, 2025
  
  e236d8fe
- chore: improve mmmu benchmark (#7000) · 4fa44d63
  Mick authored Jul 26, 2025
```
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
```
  4fa44d63