Commits · 31e9d3a5aa69b737faf2fcb1537cbfb9228bb7ea · change / sglang · GitLab

13 Sep, 2025 5 commits
- [Fix] Init mamba related memory pools with torch.zeros (#10400) · 31e9d3a5
  Binyao Jiang authored Sep 12, 2025
  
  31e9d3a5
- fix: tool parse in large streaming chunk beginning with normal content (#10397) · 6f4676ef
  Xinyuan Tong authored Sep 12, 2025
```
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  6f4676ef
- Fix the style of sgl kernel (#10398) · c9ec4cae
  Lianmin Zheng authored Sep 12, 2025
  
  c9ec4cae
- fix probs name which without temp scaling name (#9984) · 99757cc3
  narutolhy authored Sep 12, 2025
  
  99757cc3
- [Auto Sync] Update xgrammar_backend.py (20250913) (#10395) · cdddab05
  Lianmin Zheng authored Sep 12, 2025
```
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
```
  cdddab05
12 Sep, 2025 35 commits
- [router] add not implemented functions for multi model trait (#10394) · 7c5a0a1b
  Simo Lin authored Sep 12, 2025
  
  7c5a0a1b
- [HiCache] doc: update deployment in readme (#10332) · 49f169d5
  Teng Ma authored Sep 13, 2025
```
Signed-off-by: Teng Ma <sima.mt@alibaba-inc.com>
```
  49f169d5
- [HiCache] fix mooncake config in different tp size (#10377) · 7fce2fd9
  Teng Ma authored Sep 13, 2025
  
  7fce2fd9
- Support Qwen3-Next on Ascend NPU (#10379) · 16cd550c
  Even Zhou authored Sep 13, 2025
  
  16cd550c
- Benchmark: Support API_KEY without 'bearer' (#10380) · d5e2a374
  Muqi Li authored Sep 13, 2025
  
  d5e2a374
- [router] Add get and cancel method for response api (#10387) · 366043db
  Keyang Ru authored Sep 12, 2025
  
  366043db
- [router] allow one router to support different model families and serving mode (#10244) · 2f173ea0
  Simo Lin authored Sep 12, 2025
  
  2f173ea0
- Add sentencepiece to project dependencies (#10386) · 321fecab
  Mohammad Miadh Angkad authored Sep 13, 2025
  
  321fecab
- feat: add deepseek v3 fp4 ut (#10391) · 9d775b1a
  Yineng Zhang authored Sep 12, 2025
  
  9d775b1a
- Fix GPU fault issue when run dsv3 with dp mode and enable torch-compile (#10361) · 78b7465c
  kk authored Sep 13, 2025
```
Co-authored-by: wunhuang <wunhuang@amd.com>
```
  78b7465c
- [bug] fix router ci syntax error (#10390) · 07bcad7f
  Simo Lin authored Sep 12, 2025
  
  07bcad7f
- fix: exclude protobuf generated code (#10388) · 98adac8e
  Yineng Zhang authored Sep 12, 2025
  
  98adac8e
- fix: resolve gb200 image link (#10343) · cef11e9a
  Yineng Zhang authored Sep 12, 2025
  
  cef11e9a
- [Auto Sync] Update base_grammar_backend.py, llguidance_back... (20250911) (#10333) · 2269cf1e
  Lianmin Zheng authored Sep 12, 2025
```
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
```
  2269cf1e
- fix: add fast path for function call (#9023) · 151e287d
  Yi Zhang authored Sep 13, 2025
```
Co-authored-by: tazjin <mail@tazj.in>
```
  151e287d
- [router] enable sccache in ci and local build (#10099) · 8c86595c
  Simo Lin authored Sep 12, 2025
  
  8c86595c
- [router] Add Rerank Routing Logic in Regular Router (#10219) · 4634fd59
  Frank Fang authored Sep 13, 2025
  
  4634fd59
- Fix global input scale incompatible with CuTe DSL moe (#10370) · efedbe6c
  fzyzcjy authored Sep 12, 2025
  
  efedbe6c
- Fix FA4 import cause moe_fused_gate output be illegal memory (#10368) · 3a77c80b
  fzyzcjy authored Sep 12, 2025
  
  3a77c80b
- Fix chunked prefix cache for nvfp4 (#10180) · 36acd2ff
  Shu Wang authored Sep 12, 2025
```
Co-authored-by: Elfie Guo <elfieg@nvidia.com>
```
  36acd2ff
- add qwen3-next ut (#10355) · fe6cdf89
  Yi Zhang authored Sep 12, 2025
  
  fe6cdf89
- Support loading weights from remote instance (#8215) · 30d20ce8
  amysaq2023 authored Sep 12, 2025
```
Signed-off-by: Anqi Shen <amy.saq@antgroup.com>
Co-authored-by: Chayenne <74843776+zhaochenyang20@users.noreply.github.com>
```
  30d20ce8
- model: support dots.vlm1 model (#8778) · 1b1701f1
  chenge@xiaohongshu.com authored Sep 12, 2025
```
Co-authored-by: weishi <bushou@xiaohongshu.com>
Co-authored-by: Ezra-Yu <1105212286@qq.com>
Co-authored-by: Jianfei Wang <905787410@qq.com>
Co-authored-by: qianwu <wangjianfei@xiaohongshu.com>
```
  1b1701f1
- Revert add mainprocess's proctitle (#10351) · 6d403089
  ybyang authored Sep 12, 2025
  
  6d403089
- Fix Bailing MoE model bugs (#10362) · 24dc2bee
  Yuan Luo authored Sep 12, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
Co-authored-by: 羽癫 <yudian.zy@antgroup.com>
```
  24dc2bee
- Support LingV2 model (#10359) · fac07c9b
  strgrb authored Sep 12, 2025
```
Co-authored-by: 羽癫 <yudian.zy@antgroup.com>
Co-authored-by: guoyuhong <yuhong.gyh@antgroup.com>
```
  fac07c9b
- fix: resolve transfer_kv_all_layer_direct_lf_pf import error (#10360) · b3839a7f
  Yineng Zhang authored Sep 11, 2025
  
  b3839a7f
- fix the break in FlashInferFusedMoE (#10356) · 4aa39d72
  chenqianfzh authored Sep 11, 2025
```
Co-authored-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
```
  4aa39d72
- support memory_pool_host page first direct layout (#10031) · b4c2c421
  huangtingwei authored Sep 12, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  b4c2c421
- Implement Standalone gRPC Server for SGLang Python Scheduler (#10283) · 53ca1552
  Chang Su authored Sep 11, 2025
  
  53ca1552
- [router] Basic OAI Response api (#10346) · a23bdeaf
  Keyang Ru authored Sep 11, 2025
  
  a23bdeaf
- fix dual stream bug (#10352) · 27778010
  Yi Zhang authored Sep 12, 2025
  
  27778010
- model: support Apertus (#9774) · 46d8fb1c
  EduardDurech authored Sep 12, 2025
  
  46d8fb1c
- fix: flashinfer_cutlass_moe: Use max of global expert scales instead of local... · c7e85f53
  Trevor Morris authored Sep 11, 2025
```
fix: flashinfer_cutlass_moe: Use max of global expert scales instead of local for input scale (#10296)
```
  c7e85f53
- [NVIDIA] [3/N] Nvfp4 Masked Gemm: Add flashinfer grouped_gemm_nt_masked (#9199) · 3df05f4d
  Shu Wang authored Sep 11, 2025
  
  3df05f4d