Commits · be0058bc0584c44ea749e332be1432909a7228c9 · change / sglang · GitLab

20 Oct, 2025 1 commit
- [BugFix] replace the input_to_float8 used in dsv2 (#11612) · be0058bc
  Liu-congo authored Oct 20, 2025
```
Signed-off-by: Liu-congo <1502632128@qq.com>
```
  be0058bc
19 Oct, 2025 2 commits
- Change bf16 to fp8 for some gemms in attention for DeepSeek ckpt v2 (#11805) · 12eb02e9
  fzyzcjy authored Oct 19, 2025
  
  12eb02e9
- Make single-batch overlap compatible with NextN (#11804) · ce399e15
  fzyzcjy authored Oct 19, 2025
  
  ce399e15
18 Oct, 2025 4 commits
- [Refactor] move `deep_gemm_wrapper` out of `quantization` (#11784) · 5b214b50
  Cheng Wan authored Oct 17, 2025
  
  5b214b50
- Make single-batch overlap compatible with offloading (#11614) · 33e9bbec
  fzyzcjy authored Oct 18, 2025
  
  33e9bbec
- Super tiny fix CI (#11788) · dcb8f090
  fzyzcjy authored Oct 18, 2025
  
  dcb8f090
- Support casting bf16 NextN moe to fp8 (#11613) · 8af84912
  fzyzcjy authored Oct 18, 2025
  
  8af84912
17 Oct, 2025 2 commits
- Support shared experts overlap in cutlass moe (#11611) · 505329ca
  fzyzcjy authored Oct 18, 2025
  
  505329ca
- [Lint] Add `python/sglang` to ruff F401 checks and remove unused imports in files (#11685) · 62797440
  Chang Su authored Oct 17, 2025
  
  62797440
16 Oct, 2025 1 commit
- [Bugfix] Fix Qwen3/DSV3/DSV3.2 model support (#11510) · 3cceaa38
  Even Zhou authored Oct 16, 2025
  
  3cceaa38
15 Oct, 2025 1 commit
- [1/N] Introduce Mooncake Backend and Mooncake EP to Support Elastic EP (#10423) · a40229f6
  Xun Sun authored Oct 15, 2025
```
Co-authored-by: Hank Han <hanhan7630@outlook.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
```
  a40229f6
13 Oct, 2025 2 commits
- Depreate `global_server_args_dict` (#11528) · 516738b0
  Liangsheng Yin authored Oct 13, 2025
  
  516738b0
- Revert "Deprecate `global_server_args_dict`" (#11520) · 1bdd0102
  Cheng Wan authored Oct 12, 2025
  
  1bdd0102
12 Oct, 2025 1 commit
- Deprecate `global_server_args_dict` (#11331) · 1083e7e3
  Liangsheng Yin authored Oct 13, 2025
  
  1083e7e3
11 Oct, 2025 1 commit
- [BugFix] test_mla_fp8.py fails on Cublas 12.9 (#11360) · c80a96da
  Liu-congo authored Oct 11, 2025
```
Signed-off-by: Liu-congo <1502632128@qq.com>
```
  c80a96da
06 Oct, 2025 1 commit

Support DeepSeek V3.2 Exp (#11061) · efbc687c

fzyzcjy authored Oct 06, 2025


Co-authored-by: Stefan He <11166516+hebiao064@users.noreply.github.com>
Co-authored-by: Liangsheng Yin <95566987+hnyls2002@users.noreply.github.com>
Co-authored-by: Baizhou Zhang <56809903+fridge003@users.noreply.github.com>
Co-authored-by: DarkSharpness <76582120+darksharpness@users.noreply.github.com>
Co-authored-by: ZhengdQin <46387172+zhengdqin@users.noreply.github.com>
Co-authored-by: DarkSharpness <2040703891@qq.com>
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
Co-authored-by: Zhengda Qin <zhengdqin@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: HAI <hixiao@gmail.com>
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>

efbc687c

02 Oct, 2025 5 commits
- Tiny cleanup deepseek_v2.py (#11163) · b65db028
  fzyzcjy authored Oct 02, 2025
  
  b65db028
- Support single batch overlap (#10422) · 5e786cca
  fzyzcjy authored Oct 02, 2025
  
  5e786cca
- Support dispatch low latency (#10263) · 0b9dfba7
  fzyzcjy authored Oct 02, 2025
```
Co-authored-by: Kaixi Hou <4001424+kaixih@users.noreply.github.com>
```
  0b9dfba7
- Fuse quantize and rope in trtllm_mla MTP (#10779) · f35def86
  fzyzcjy authored Oct 02, 2025
  
  f35def86
- Fix DeepSeek chunked prefill memory issue (#11149) · 44b1fbe2
  fzyzcjy authored Oct 02, 2025
  
  44b1fbe2
23 Sep, 2025 1 commit
- [Feature] Add MLAProcess for DeepSeek MLA on NPU (#10130) · d27a6f70
  Even Zhou authored Sep 23, 2025
  
  d27a6f70
22 Sep, 2025 1 commit

[Auto Sync] Update deepseek_v2.py (20250922) (#10717) · f67d1f45

Yineng Zhang authored Sep 21, 2025


Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Michael Granado <mgranado@together.ai>

f67d1f45

20 Sep, 2025 2 commits
- fix: remove awq_dequantize deps (#10686) · 7c876de7
  Yineng Zhang authored Sep 20, 2025
  
  7c876de7
- [Auto Sync] Update deepseek_v2.py (20250920) (#10683) · b17e67df
  Yineng Zhang authored Sep 19, 2025
```
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
```
  b17e67df
17 Sep, 2025 1 commit
- enable prefix cache with dp (#10459) · 124097fc
  Shu Wang authored Sep 16, 2025
  
  124097fc
16 Sep, 2025 1 commit
- support using fa4 on deepseek on blackwell (#9928) · a2f7218a
  cicirori authored Sep 17, 2025
  
  a2f7218a
15 Sep, 2025 3 commits
- Fix trtllm_moe wrong correction bias (#10440) · 059c13de
  fzyzcjy authored Sep 15, 2025
  
  059c13de
- Refactor TopK to ensure readability and extensibility (#9338) · 4844fac9
  Cheng Wan authored Sep 14, 2025
  
  4844fac9
- Fix correction bias undefined behavior for nvfp4 models (#10426) · 258d02c8
  fzyzcjy authored Sep 15, 2025
  
  258d02c8
14 Sep, 2025 2 commits
- Support offloading in fp8 (#9948) · fa46e2bd
  fzyzcjy authored Sep 14, 2025
  
  fa46e2bd
- [2/2] Speed up prefill mla attention concat (#10157) · b047b553
  fzyzcjy authored Sep 14, 2025
  
  b047b553
12 Sep, 2025 2 commits
- Fix chunked prefix cache for nvfp4 (#10180) · 36acd2ff
  Shu Wang authored Sep 12, 2025
```
Co-authored-by: Elfie Guo <elfieg@nvidia.com>
```
  36acd2ff
- [NVIDIA] [3/N] Nvfp4 Masked Gemm: Add flashinfer grouped_gemm_nt_masked (#9199) · 3df05f4d
  Shu Wang authored Sep 11, 2025
  
  3df05f4d
10 Sep, 2025 1 commit
- Add tests to AMD CI for MI35x (#9662) · 91b3555d
  Hubert Lu authored Sep 10, 2025
```
Co-authored-by: Sai Enduri <saimanas.enduri@amd.com>
```
  91b3555d
09 Sep, 2025 3 commits
- Revert "Revert the changes on NCCL symmetric memory" (#10238) · 4582931a
  Lianmin Zheng authored Sep 09, 2025
  
  4582931a
- Revert the changes on NCCL symmetric memory (#10210) · d352c29a
  Lianmin Zheng authored Sep 09, 2025
```
Co-authored-by: Yineng Zhang <me@zhyncs.com>
```
  d352c29a
- Cleaning codes for speculative attention mode (#10149) · 8ad700f7
  Baizhou Zhang authored Sep 08, 2025
  
  8ad700f7
08 Sep, 2025 2 commits
- Add speculator attention backend switch (#9981) · 8c5930f0
  cicirori authored Sep 08, 2025
  
  8c5930f0
- Fix run time error in dsv3-fp8 model on mi35x (#10104) · 400d3b97
  kk authored Sep 08, 2025
```
Co-authored-by: wunhuang <wunhuang@amd.com>
Co-authored-by: HaiShaw <hixiao@gmail.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
```
  400d3b97