Commits · 901ab758ecc47ea900632d1393084029413164fc · change / sglang

"vscode:/vscode.git/clone" did not exist on "8e46d97cd894998fb944b5021cdd4f7d6bfcd39c"

05 Aug, 2025 10 commits
- chore: upgrade transformers 4.55.0 (#8823) · 901ab758
  Yineng Zhang authored Aug 05, 2025
```
Co-authored-by: hebiao064 <hebiaobuaa@gmail.com>
```
  901ab758
- GLM-4.5 and GLM-4.5-Air both support (#8804) · a4b0d5c9
  Yuxuan Zhang authored Aug 05, 2025
  
  a4b0d5c9
- feat: add trtllm-gen mha from direct call (#8782) · 40e3b2be
  eigen authored Aug 05, 2025
```
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
```
  40e3b2be
- Revert "[NVIDIA]Fix local_num_experts for EP (#8779)" (#8797) · 5e91fed1
  Yineng Zhang authored Aug 04, 2025
  
  5e91fed1
- [feat] Add detail in image_data (#8596) · 873f384a
  Yuhao Yao authored Aug 05, 2025
  
  873f384a
- [NVIDIA]Fix local_num_experts for EP (#8779) · b01eeb80
  Shu Wang authored Aug 05, 2025
  
  b01eeb80
- chore: upgrade flashinfer v0.2.9 (#8780) · 1ea94d3b
  Yineng Zhang authored Aug 04, 2025
  
  1ea94d3b
- [PD] Refactor parallel sizes and add pp support for mooncake (#8571) · d98a4913
  Shangming Cai authored Aug 05, 2025
```
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
```
  d98a4913
- Support OCP MXFP4 quantization on AMD GPUs (#8255) · d4bf5a85
  kk authored Aug 05, 2025
```
Co-authored-by: wunhuang <wunhuang@amd.com>
Co-authored-by: Hubert Lu <Hubert.Lu@amd.com>
```
  d4bf5a85
- [Fix] Fix several issues preventing gemma3n LoRA support. (#8776) · 7cb20754
  Lifu Huang authored Aug 04, 2025
  
  7cb20754
04 Aug, 2025 9 commits
- [NVIDIA] Fix breakage of using trtllm-gen fp8 moe (#8773) · 6d0646da
  Kaixi Hou authored Aug 04, 2025
  
  6d0646da
- [bugfix] Fix typo in modelopt quant: 'FusedMoE' object has no attribute 'local_num_experts' (#8768) · 9bd4872a
  Trevor Morris authored Aug 04, 2025
  
  9bd4872a
- [NVIDIA] Add Low Latency NVFP4 decode kernels from Flashinfer (#8552) · 915140fd
  azhurkevich authored Aug 04, 2025
```
Co-authored-by: Cheng Wan <cwan@x.ai>
```
  915140fd
- [bugfix] fix import path in HiCacheController (#8749) · 36fc9260
  Baron Liu authored Aug 04, 2025
  
  36fc9260
- [CI] Ascend NPU CI enhancement (#8294) · fee0ab0f
  Even Zhou authored Aug 04, 2025
```
Co-authored-by: ronnie_zheng <zl19940307@163.com>
```
  fee0ab0f
- Rename lora_path to lora_id in batches (#8437) · f2d68ded
  Baizhou Zhang authored Aug 03, 2025
  
  f2d68ded
- Fix bug of refactoring TopKOutput in w4afp8 (#8745) · 3b87a9e8
  Yuan Luo authored Aug 04, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  3b87a9e8
- Replace torch.jit.script with torch.compile in get_masked_input_and_mask to... · f024795e
  YyWangCS authored Aug 04, 2025
```
Replace torch.jit.script with torch.compile in get_masked_input_and_mask to fix benchmark underreporting (#8733)
```
  f024795e
- [MoE] Enable `renormalize=False` in Triton kernels (#8735) · b102353f
  Cheng Wan authored Aug 03, 2025
  
  b102353f
03 Aug, 2025 21 commits
- fix args typo in memory_pool_host (#8662) · 76ba5bbe
  huangtingwei authored Aug 04, 2025
  
  76ba5bbe
- Fix the missing 'lof' choice of --schedule-policy server args (#7114) · ed6f7597
  Yingchun Lai authored Aug 04, 2025
  
  ed6f7597
- feat: support cutlass_moe_fp8 kernel for fusedmoe in sm90 (#8678) · e67276ec
  tql.99 authored Aug 04, 2025
  
  e67276ec
- Fix triton kernels topk with keyword arguments (#8732) · 0242bb9c
  Ke Bao authored Aug 04, 2025
  
  0242bb9c
- use fp32 for e_score_correction_bias in GLM-4.5 (#8729) · 760286e3
  Yuxuan Zhang authored Aug 04, 2025
  
  760286e3
- [RL] fix update weight for FusedMoE with EP (#8676) · 3435a24e
  Zilin Zhu authored Aug 04, 2025
  
  3435a24e
- feat: Support DP Attention for step3_vl (#8699) · 00da9065
  yhyang201 authored Aug 03, 2025
  
  00da9065
- chore: bump v0.4.10.post2 (#8727) · 8cd34458
  Yineng Zhang authored Aug 03, 2025
  
  8cd34458
- [DP] fix the compatibility issue between DP attention and `--attention-backend triton` (#8723) · 0e0eef00
  Cheng Wan authored Aug 03, 2025
  
  0e0eef00
- [CUDA Graph] save cuda graph memory by using next_token_logits_buffer (#8579) · cb099d20
  Cheng Wan authored Aug 03, 2025
  
  cb099d20
- Save cuda graph memory for fa3 (#8567) · 7a913301
  Cheng Wan authored Aug 03, 2025
  
  7a913301
- [Improvements] Merge health check route (#8444) · 6f9baf10
  ybyang authored Aug 03, 2025
```
Signed-off-by: ybyang <ybyang7@iflytek.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
Co-authored-by: Kan Wu <wukanustc@gmail.com>
```
  6f9baf10
- feat: Add new moe triton for NVIDIA RTX 6000 Ada (#8547) · a31b7a70
  Jasper James authored Aug 03, 2025
  
  a31b7a70
- [fix] Fix divide by zero error for llama4. (#8683) · 7ed8e51b
  Varun Vinayak Shenoy authored Aug 03, 2025
  
  7ed8e51b
- Do layernorm before allgather for DP attention (#8631) · 32f28154
  Trevor Morris authored Aug 03, 2025
  
  32f28154
- [feat] support minimum token load balance in dp attention (#7379) · f7b2853f
  Guanhua Wang authored Aug 03, 2025
  
  f7b2853f
- HiCache storage, style change and bug fix (#8719) · b0add2da
  Zhiqiang Xie authored Aug 03, 2025
  
  b0add2da
- Reduce memory accumulation in long-running server (#8306) · 0305c505
  Wenxuan Tan authored Aug 03, 2025
```
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
```
  0305c505
- Support limiting max loaded loras in CPU. (#8650) · 8675bdf2
  Lifu Huang authored Aug 03, 2025
  
  8675bdf2
- [hotfix] fix mixtral with tensor-level compressed-tensor quantization (#8721) · a437aa99
  Cheng Wan authored Aug 02, 2025
  
  a437aa99
- Tiny fix CI pytest error (#8524) · 0e612dbf
  fzyzcjy authored Aug 03, 2025
  
  0e612dbf