Commits · d904959233bb21f5bb713ac3da46da616160d3f3 · change / sglang · GitLab

31 Jul, 2025 6 commits
- Support l3 cache (mooncake store) for hiradix cache (#7211) · d9049592
  huangtingwei authored Jul 31, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
Co-authored-by: AniZpZ <zhuangsen.zp@antgroup.com>
Co-authored-by: zuoyuan <zhangzuo21@mails.tsinghua.edu.cn>
Co-authored-by: @wangyueneng.wyn <wangyueneng.wyn@antgroup.com>
Co-authored-by: JinYan Su <jinyansu792@gmail.com>
```
  d9049592
- fix incorrect increase of hit count (#8533) · 26c8a310
  huangtingwei authored Jul 31, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  26c8a310
- [bugfix] Fix 2 minor bugs in the hicache storage layer (#8404) · 5963e505
  yi wang authored Jul 31, 2025
  
  5963e505
- Bug: Fix google gemma3n-mm audio input not working bug (#8365) · 59aab76f
  Binyao Jiang authored Jul 30, 2025
  
  59aab76f
- Bump transfomers to 4.54.1 to fix Gemma cache issue. (#8541) · 67e53b16
  Lifu Huang authored Jul 30, 2025
  
  67e53b16
- Add hf3fs support for hicache storage (based on #7704) (#7280) · 29980334
  pansicheng authored Jul 31, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  29980334
30 Jul, 2025 5 commits
- Revert "Fix the input tools format and history tool_calls in OpenAI API (#6556)" (#8584) · a79a5d70
  Chang Su authored Jul 30, 2025
  
  a79a5d70
- [Model] Add support for Arcee Foundational Model (#8154) · ec5f9442
  Adarsh Shirawalmath authored Jul 30, 2025
  
  ec5f9442
- Update cutlass_moe.py (#8545) · e3f08c77
  Elfie Guo authored Jul 29, 2025
  
  e3f08c77
- feature(pd-hicache): Prefill instances support reusing the RemoteStorage Cache via HiCache. (#8516) · 2fbb754e
  hzh0425 authored Jul 30, 2025
```
Co-authored-by: Shangming Cai <csmthu@gmail.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  2fbb754e
- feat(hicache): support file backend reading directory config form env. (#8498) · a85ebf50
  hzh0425 authored Jul 30, 2025
  
  a85ebf50
29 Jul, 2025 9 commits
- Support EPLB in FusedMoE (#8448) · 9effeb5b
  Cheng Wan authored Jul 29, 2025
  
  9effeb5b
- fix: temporarily disable cuda-ipc for mm data tensor (#8431) · 1992ef9b
  Mick authored Jul 30, 2025
```
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
```
  1992ef9b
- Split the scheduler into multiple mixin classes to reduce the file size (#8483) · a4c3b121
  Lianmin Zheng authored Jul 29, 2025
  
  a4c3b121
- Update cutlass_moe.py (#8535) · 4d16c88b
  Elfie Guo authored Jul 29, 2025
  
  4d16c88b
- chore: bump v0.4.9.post6 (#8517) · 6478831b
  Yineng Zhang authored Jul 29, 2025
  
  6478831b
- Fix incorrect KV cache allocation for MTP models. (#8482) · fb16fbaf
  Lifu Huang authored Jul 28, 2025
```
Co-authored-by: Stefan He <hebiaobuaa@gmail.com>
```
  fb16fbaf
- Support colocating requests (#7973) · 0ce84c82
  fzyzcjy authored Jul 29, 2025
  
  0ce84c82
- Tiny add warnings for DeepEP when it is suboptimal (#8426) · 59d0bf01
  fzyzcjy authored Jul 29, 2025
  
  59d0bf01
- Reduce memory usage for fp4 moe (#8413) · 7df2c0c2
  fzyzcjy authored Jul 29, 2025
  
  7df2c0c2
28 Jul, 2025 17 commits
- chore: add glm 4.5 fp8 tp4 config (#8480) · 8240a6b0
  Yineng Zhang authored Jul 28, 2025
  
  8240a6b0
- chore: add glm4 fp8 tp8 config (#8478) · 3a04aa4b
  Yineng Zhang authored Jul 28, 2025
  
  3a04aa4b
- Fix DEEPEP BF16 compatibility for Deepseek Style model like GLM 4.5 (#8469) · 74e7e457
  Stefan He authored Jul 28, 2025
```
Co-authored-by: Minglei Zhu <mingleizhu1122@gmail.com>
```
  74e7e457
- [3/N] MoE Refactor: Simplify DeepEP Output (#8421) · 9c138a04
  Cheng Wan authored Jul 28, 2025
  
  9c138a04
- Fix parsing ChatCompletionMessage (#7273) · c8f549d9
  Timofey authored Jul 28, 2025
```
Co-authored-by: Timofey K <timosha1113@gmail.com>
```
  c8f549d9
- [NVIDIA] Change to use `num_local_experts` (#8453) · 134fa43e
  Kaixi Hou authored Jul 28, 2025
  
  134fa43e
- fix: update dep (#8467) · ccfe52a0
  Yineng Zhang authored Jul 28, 2025
  
  ccfe52a0
- feat: throttle requests at scheduler based on --max_queued_requests (#7565) · 747dd450
  harrisonlimh authored Jul 28, 2025
  
  747dd450
- fix:reorder topk experts to ensure shared expert replaces minimal score (#8125) · a9dd3ec3
  erictanjn authored Jul 28, 2025
  
  a9dd3ec3
- chore: bump v0.4.9.post5 (#8458) · 45bc170b
  Yineng Zhang authored Jul 28, 2025
  
  45bc170b
- fix GLM4_MOE launch with compressed_tensor quant model (#8456) · 25f73c6c
  Minglei Zhu authored Jul 28, 2025
  
  25f73c6c
- GLM-4.5 Model Support Follow-up (#8445) · 581e7dcb
  Binyao Jiang authored Jul 27, 2025
  
  581e7dcb
- 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
- [PD] Fix abort_request for PD disaggregation (#8352) · 2fd5c704
  Shangming Cai authored Jul 28, 2025
```
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
Co-authored-by: ybyang <10629930+whybeyoung@users.noreply.github.com>
```
  2fd5c704
- chore: bump transformer to 4.54.0 (#8416) · 4ad97370
  Stefan He authored Jul 27, 2025
```
Co-authored-by: Binyao Jiang <byjiang1996@gmail.com>
Co-authored-by: Lifu Huang <lifu.hlf@gmail.com>
```
  4ad97370
- [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
- bugfix: Fix XGrammar backend to use model's EOS tokens for constrained generation (#8422) · dd487e55
  Chang Su authored Jul 27, 2025
  
  dd487e55
27 Jul, 2025 3 commits
- 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