Commits · ed91561f7972e280a45df7420c40858de35c49e7 · change / sglang

"vscode:/vscode.git/clone" did not exist on "579210d21411cf819b90839f72d89116f7df7488"

12 Mar, 2025 4 commits
- upgrade sgl-kernel 0.0.4.post3 (#4334) · ed91561f
  Yineng Zhang authored Mar 12, 2025
  
  ed91561f
- Remove vllm ops scaled fp8 quant and accelerate per token quant by 20-28% (#4215) · e0917e6b
  Stefan He authored Mar 12, 2025
```
Co-authored-by: Stefan He <bhe@linkedin.com>
```
  e0917e6b
- Add A800 tuning configs for DeepSeek R1/V3 channel-wise INT8 (#4323) · 7140ba35
  lambert0312 authored Mar 12, 2025
  
  7140ba35
- unify is_cuda and is_hip (#4321) · d1da58e2
  Yineng Zhang authored Mar 11, 2025
  
  d1da58e2
11 Mar, 2025 8 commits
- upgrade flashinfer 0.2.3 (#4317) · 1cf63485
  Yineng Zhang authored Mar 11, 2025
```
Co-authored-by: qingquansong <qsong@linkedin.com>
```
  1cf63485
- refactor: move image processors to separate files (#4229) · ff2ce0b8
  Mick authored Mar 12, 2025
  
  ff2ce0b8
- Add H20 tuning configs support DeepSeek V3/R1 INT8(block-wise) (#4220) · 0f2a2e3c
  Ximingwang-09 authored Mar 12, 2025
```
Co-authored-by: ximing.wxm <ximing.wxm@antgroup.com>
```
  0f2a2e3c
- [AMD] Fix rocm sgl-kernel missing modules error (#4311) · 690e1f23
  yigex authored Mar 12, 2025
```
Co-authored-by: yiakwy-xpu-ml-framework-team <leiwang2@amd.com>
```
  690e1f23
- Add A100 tuning configs for DeepSeek R1/V3 channel-wise INT8 (#4287) · 6a02b32d
  yych0745 authored Mar 11, 2025
```
Co-authored-by: HandH1998 <1335248067@qq.com>
```
  6a02b32d
- linear support deepgemm (#4199) · dce303e2
  lukec authored Mar 11, 2025
```
Co-authored-by: yinfan98 <1106310035@qq.com>
```
  dce303e2
- update sgl-kernel 0.0.4.post2 (#4291) · 4d27eb9a
  Yineng Zhang authored Mar 11, 2025
  
  4d27eb9a
- Add A800 tuning configs support DeepSeek V3/R1 BF16 and INT8(block-wise) (#4136) · d3ecd632
  lambert0312 authored Mar 11, 2025
  
  d3ecd632
10 Mar, 2025 11 commits
- upgrade xgrammar 0.1.15 (#4275) · e187a3d5
  Yineng Zhang authored Mar 10, 2025
  
  e187a3d5
- Amd test fp8 (#4261) · 2ac189ed
  HandH1998 authored Mar 11, 2025
  
  2ac189ed
- Test no vllm custom allreduce (#4256) · 5a6400ee
  Lianmin Zheng authored Mar 10, 2025
  
  5a6400ee
- Fix quantization and nightly tests (#4258) · 00d25a7f
  Lianmin Zheng authored Mar 10, 2025
  
  00d25a7f
- fix the input_ids is None error (#4144) · ac698850
  shimin authored Mar 10, 2025
  
  ac698850
- Simplify tests & Fix trtllm custom allreduce registration (#4252) · aa957102
  Lianmin Zheng authored Mar 10, 2025
  
  aa957102
- [Bug fixed] fixed the crash when enable the dp-attention on the single card (#3958) · 4455b26e
  DavidChan authored Mar 10, 2025
  
  4455b26e
- Clean up fp8 support (#4230) · e8a69e4d
  Lianmin Zheng authored Mar 09, 2025
  
  e8a69e4d
- Auto balance CI tests (#4238) · fbd56002
  Lianmin Zheng authored Mar 09, 2025
  
  fbd56002
- Minor style fix for sgl-kernel (#4243) · 730d084f
  Lianmin Zheng authored Mar 09, 2025
  
  730d084f
- Revert "Check eagle server args" (#4242) · 4a05bdfa
  Lianmin Zheng authored Mar 09, 2025
  
  4a05bdfa
09 Mar, 2025 6 commits
- Check eagle server args (#4217) · 34c88987
  Ying Sheng authored Mar 09, 2025
  
  34c88987
- Apply sgl w8a8 fp8 kernel (#3148) · 0dd6cda2
  HandH1998 authored Mar 09, 2025
  
  0dd6cda2
- Support nextn for flashinfer mla attention backend (#4218) · 9fb48f95
  Baizhou Zhang authored Mar 09, 2025
  
  9fb48f95
- use sgl-kernel 0.0.4 (#4224) · 89ccb533
  Yineng Zhang authored Mar 08, 2025
  
  89ccb533
- Lazily import lora backends (#4225) · 1361ab9e
  Lianmin Zheng authored Mar 08, 2025
  
  1361ab9e
- Rename files in sgl kernel to avoid nested folder structure (#4213) · 8abf74e3
  Lianmin Zheng authored Mar 08, 2025
```
Co-authored-by: zhyncs <me@zhyncs.com>
```
  8abf74e3
08 Mar, 2025 4 commits
- Fix bench_serving flush cache not recognizing OPENAI_API_KEY (#4181) · 0fe7c13b
  Mingshan authored Mar 08, 2025
```
Signed-off-by: Mingshan <git@brighill.com>
```
  0fe7c13b
- lazy import attn backends (#4200) · 08c4d764
  Lianmin Zheng authored Mar 08, 2025
  
  08c4d764
- [EAGLE] many fixes for eagle (#4195) · d4017a6b
  Lianmin Zheng authored Mar 07, 2025
```
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: Sehoon Kim <sehoon@x.ai>
```
  d4017a6b
- New clang format for sgl kernel (#4194) · d052f4c8
  Lianmin Zheng authored Mar 07, 2025
  
  d052f4c8
07 Mar, 2025 6 commits
- Fix eagle hang issue for max_new_tokens=1 (#4185) · 20c81199
  Ke Bao authored Mar 08, 2025
  
  20c81199
- Revert "ROCm: Flex Attention Enablement with custom backends (#4178)" (#4186) · eb61f5c9
  Yineng Zhang authored Mar 07, 2025
  
  eb61f5c9
- ROCm: Flex Attention Enablement with custom backends (#4178) · 0beea450
  HAI authored Mar 07, 2025
```
Co-authored-by: linsun12 <linsun12@amd.com>
```
  0beea450
- Memory pool fix for upstream change about eagle (#4170) · 9376ac36
  Zhiqiang Xie authored Mar 07, 2025
  
  9376ac36
- [Feature] DeepSeek V3/R1 INT8 Quantization (channel-wise) (#3888) · c7f25446
  HandH1998 authored Mar 07, 2025
```
Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
Co-authored-by: b0urnee <2769086541@qq.com>
```
  c7f25446
- Add Support for Qwen2-VL Multi-modal Embedding Models (#3694) · 361971b8
  uylnap authored Mar 07, 2025
  
  361971b8
06 Mar, 2025 1 commit
- ROCm: enable trillion-parameter MoE models with INT4-FP8 single node (#4152) · 13bc39c5
  HAI authored Mar 06, 2025
  
  13bc39c5