Commits · fd71b11b1d96d385b09cb79c91a36f1f01293639 · change / sglang · GitLab

27 Aug, 2025 10 commits
- move is_sm90_supported/is_sm100_supported to python/sglang/srt/utils.py (#9679) · fd71b11b
  Lianmin Zheng authored Aug 27, 2025
  
  fd71b11b
- fix mooncake store mla zero copy meta (#9678) · ae7428a8
  huangtingwei authored Aug 27, 2025
  
  ae7428a8
- fix: HiRadixCache: fix prefetch completion race (#9397) · a3aee7c3
  Pablo Iyu Guerrero authored Aug 27, 2025
  
  a3aee7c3
- support cuda 13.0 and trtllm kernel by Aug 25 2025 (#9495) · 79e6a8a6
  Rain Jiang authored Aug 26, 2025
  
  79e6a8a6
- Add A100 fused MoE kernel configs for Dpsk (#9677) · 8f7b1c31
  ehuaa authored Aug 27, 2025
  
  8f7b1c31
- Support DeepSeek-V3.1 tool call (#9446) · b9683be6
  Xu Wenqing authored Aug 27, 2025
```
Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
```
  b9683be6
- [docs] Instructions for bench_serving.py (#9071) · a85363c1
  yhyang201 authored Aug 27, 2025
```
Co-authored-by: Mick <mickjagger19@icloud.com>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
Co-authored-by: zhaochenyang20 <zhaochen20@outlook.com>
Co-authored-by: zhaochenyang20 <zhaochenyang20@gmail.com>
Co-authored-by: Yineng Zhang <me@zhyncs.com>
```
  a85363c1
- feat: (chat-template matching) enhance multimodal model detection with config.json (#9597) · b21fdd53
  Kevin Tuan authored Aug 27, 2025
  
  b21fdd53
- refactor(hicache): Introduce generic HiCacheStorageConfig for improved... · c04c17ed
  hzh0425 authored Aug 27, 2025
```
refactor(hicache): Introduce generic HiCacheStorageConfig for improved configuration management (#9555)
Co-authored-by: Teng Ma <805522925@qq.com>
```
  c04c17ed
- chore: enhance bench_serving for vlms with a new dataset of configurable image... · 16a6d21b
  Mick authored Aug 27, 2025
```
chore: enhance bench_serving for vlms with a new dataset of configurable image count and resolution (#9583)
Co-authored-by: yhyang201 <yhyang201@gmail.com>
```
  16a6d21b
26 Aug, 2025 23 commits
- [RL] fix register the same ops multiple times (#9564) · a530b3ff
  Stefan He authored Aug 26, 2025
  
  a530b3ff
- Fix FA3 swa spec verify topk>1 (#9658) · 603b3446
  Ke Bao authored Aug 27, 2025
  
  603b3446
- add `response_format` support for `completion` API (#9665) · b6c14ec0
  cicirori authored Aug 27, 2025
  
  b6c14ec0
- HiCache Storage fix host memory leak (#9648) · 43de1d73
  Zhiqiang Xie authored Aug 26, 2025
  
  43de1d73
- BugFix(hicache): Fix host indices out of bound error (#9637) · 79ce3688
  hzh0425 authored Aug 27, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  79ce3688
- Install py-spy by default for containers for easier debugging (#9649) · 44ffe2cb
  fzyzcjy authored Aug 27, 2025
  
  44ffe2cb
- [doc] add kimik2 --tool-call-parser (#9647) · 1a0896e9
  Xiaotong Jiang authored Aug 26, 2025
  
  1a0896e9
- [router] add token bucket rate limiter (#9656) · 90313fb0
  Chang Su authored Aug 26, 2025
  
  90313fb0
- [router] address worker load tracking consistency (#9523) · 3578eb1e
  Simo Lin authored Aug 26, 2025
```
Co-authored-by: fzyzcjy <5236035+fzyzcjy@users.noreply.github.com>
```
  3578eb1e
- Fix kimi k2 function calling format (#9606) · 0936c766
  Xiaotong Jiang authored Aug 26, 2025
  
  0936c766
- fix: allow user to specify function as role (#9635) · 0ef583b7
  GavinZhu-GMI authored Aug 26, 2025
  
  0ef583b7
- Add reasoning_effort param in TiktokenTokenizer.apply_chat_template (#9630) · f7881a27
  Liu Shaohui authored Aug 26, 2025
```
Co-authored-by: Shaohui Liu <liushaohui3@xiaomi.com>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
```
  f7881a27
- [docs] Update README with additional highlights and resources for SGLang x AMD SF Meetup (#9640) · fdff3167
  Mingyi authored Aug 26, 2025
  
  fdff3167
- Fix lint for router (#9636) · cbc0e4d7
  Stefan He authored Aug 26, 2025
  
  cbc0e4d7
- model: Support nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 (#9301) · 4cd08dc5
  Netanel Haber authored Aug 26, 2025
  
  4cd08dc5
- [new feat] ascend backend support fia fusion kernel (#8328) · f92b729d
  ZhengdQin authored Aug 26, 2025
```
Co-authored-by: Even Zhou <even.y.zhou@outlook.com>
```
  f92b729d
- [router] add ut for mistral, llama, pythonic, and streaming tool parser (#9632) · e2e378ca
  Simo Lin authored Aug 25, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  e2e378ca
- [router] add llama tool parser (#9629) · dc1decc6
  Simo Lin authored Aug 25, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  dc1decc6
- [router] add pythonic parser (#9628) · 03680f33
  Simo Lin authored Aug 25, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  03680f33
- [router] add qwen tool parser (#9623) · d4c5e534
  Simo Lin authored Aug 25, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  d4c5e534
- [router] add mistral tool parser (#9622) · 817c62a0
  Simo Lin authored Aug 25, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
```
  817c62a0
- Improve bench_one_batch_server script (#9608) · 0ff72419
  Liangsheng Yin authored Aug 26, 2025
```
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  0ff72419
- [Fix] HiCache Bugfix & Mooncake Error Handling Enhance (#8901) · 80dc76e1
  ykwd authored Aug 26, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  80dc76e1
25 Aug, 2025 7 commits
- [docs] Refactor, remove compiled results and add gpt-oss (#9613) · 9b08d975
  Chayenne authored Aug 25, 2025
```
Co-authored-by: zhaochenyang20 <zhaochenyang20@gmail.com>
```
  9b08d975
- Fix Harmony reasoning parser for and auto-separation for gpt-oss models (#9190) · a0a77d93
  Jonas authored Aug 26, 2025
```
Co-authored-by: Chang Su <chang.s.su@oracle.com>
Co-authored-by: Chayenne <zhaochen20@outlook.com>
Co-authored-by: zhaochenyang20 <zhaochenyang20@gmail.com>
Co-authored-by: minleminzui <2969413251@qq.com>
Co-authored-by: maocheng23 <maocheng@berkeley.edu>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
```
  a0a77d93
- Fix GLM45v launch server cuda torch compile bug (#9554) · 24a8cee6
  Binyao Jiang authored Aug 25, 2025
  
  24a8cee6
- Fix GLM45 tool call multi-turn bug (#9500) · 3affa9dc
  Binyao Jiang authored Aug 25, 2025
  
  3affa9dc
- [Performance] Batch Send from Tokenizer Manager. (#9436) · ea0696b9
  Sundara Raman Ramachandran authored Aug 25, 2025
  
  ea0696b9
- [Doc] add LWS(LeaderWorkerSet) use case in sgl-router README (#9568) · 3aec3d4f
  Bruce-x-1997 authored Aug 25, 2025
```
Co-authored-by: bruce.xu <bruce.xu@gmicloud.ai>
```
  3aec3d4f
- chore: bump v0.5.1.post2 (#9592) · e3e97a12
  Yineng Zhang authored Aug 25, 2025
  
  e3e97a12