Commits · d70a16625dc74d9517641aa82f4ae7367854da96 · OpenDAS / vllm_cscc

21 Aug, 2025 7 commits

[Performance] V1 Pooling Models E2E Performance Optimization (#23162) · d70a1662
wang.yuqi authored Aug 21, 2025
```
Signed-off-by: wang.yuqi <noooop@126.com>
```
d70a1662
[Model] Add LFM2 architecture (#22845) · 2e2000f3
Paul Pak authored Aug 21, 2025
```
Signed-off-by: Paul Pak <paulpak58@gmail.com>
```
2e2000f3
[Bugfix] Fix extra whitespace in strings caused by newline (#23272) · 0c31e28e
Cyrus Leung authored Aug 21, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
0c31e28e

[Model][VLM] Support R-4B Model (#23246) · 655a09f6

杨奇(yann qi) authored Aug 21, 2025

Signed-off-by: yannqi <yannqi@qq.com>
Signed-off-by: 杨奇(yann qi) <51905299+yannqi@users.noreply.github.com>
Signed-off-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Co-authored-by: yannqiyang <yannqiyang@tencent.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>

655a09f6

[V1][Mamba1] - Full CUDA and Piecewise CUDA Graphs Support (#23035) · 3663870c

Asaf Joseph Gardin authored Aug 21, 2025


Signed-off-by: asafg <asafg@ai21.com>
Signed-off-by: asafg <39553475+Josephasafg@users.noreply.github.com>
Co-authored-by: asafg <asafg@ai21.com>

3663870c

[CI/Build] Split out mm processor tests (#23260) · 2461d9e5
Cyrus Leung authored Aug 21, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
2461d9e5
[CPU] Refactor CPU W8A8 scaled_mm (#23071) · 7be5d113
Li, Jiang authored Aug 21, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
7be5d113

20 Aug, 2025 14 commits
- [EP] Add logging for experts map (#22685) · 4b795020
  22quinn authored Aug 20, 2025
```
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
Co-authored-by: Simon Mo <simon.mo@hey.com>
```
  4b795020
- [Feature] use --eplb_config to set eplb param (#20562) · 4fbda0b2
  rongfu.leng authored Aug 21, 2025
```
Signed-off-by: rongfu.leng <rongfu.leng@daocloud.io>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: rongfu.leng <lenronfu@gmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  4fbda0b2
- [Kernel/Quant] Remove the original marlin format and qqq (#23204) · 0cdbf5e6
  Michael Goin authored Aug 20, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  0cdbf5e6
- Small fix for Command-A-Vision (#23268) · ebe56a00
  dongluw authored Aug 20, 2025
```
Signed-off-by: donglu <donglu@cohere.com>
```
  ebe56a00
- Migrate Mistral3ImagePixelInputs to TensorSchema (#21945) · c4477f55
  Benji Beck authored Aug 20, 2025
```
Signed-off-by: Benji Beck <benjibeck@meta.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  c4477f55
- [CLI][Doc] Formalize `--mm-encoder-tp-mode` (#23190) · 5efd6905
  Cyrus Leung authored Aug 20, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  5efd6905
- [Kernel] CUTLASS MoE FP8: Integrate cuda moe permute/unpermute (#23045) · b17109be
  shixianc authored Aug 20, 2025
```
Signed-off-by: Shixian Cui <shixian@amazon.com>
```
  b17109be
- [Bugfix] Ensure correctness of HCXVision processing (#23254) · 44492358
  Cyrus Leung authored Aug 20, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  44492358
- [Model] Improve olmo and olmo2 (#23228) · c6d80a7a
  Jee Jee Li authored Aug 20, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  c6d80a7a
- [Model][V1] Support Ernie MTP (#22169) · 7cd17e22
  xyxinyang authored Aug 20, 2025
```
Signed-off-by: zhouchong <zhouchong03@baidu.com>
Co-authored-by: zhouchong <zhouchong03@baidu.com>
```
  7cd17e22
- [Bugfix] Ensure correctness of Cohere2Vision processing (#23245) · 68fcd3fa
  Cyrus Leung authored Aug 20, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  68fcd3fa
- [Model] Support deepseek with eagle (#21086) · 83e69a09
  Xin Yang authored Aug 20, 2025
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
```
  83e69a09
- [Model] use autoWeightsLoader for gptoss (#22446) · 103f1ec8
  Calvin Chen authored Aug 20, 2025
```
Signed-off-by: calvin chen <wen.chen@dynamia.ai>
```
  103f1ec8
- [Doc] Update V1 status of various pooling models (#23189) · 64ab3c72
  Cyrus Leung authored Aug 20, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  64ab3c72
19 Aug, 2025 13 commits
- [NVIDIA] Add SM100 Flashinfer Cutlass MoE fp8 backend (#22357) · a38b8af4
  amirkl94 authored Aug 20, 2025
```
Signed-off-by: Amir Klein <203507526+amirkl94@users.noreply.github.com>
```
  a38b8af4
- [Bugfix] Fix accuracy issue when using flashinfer cutlass moe, TP=1 and modelopt. (#23125) · b94faf9d
  bnellnm authored Aug 19, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  b94faf9d
- Migrate LlavaOnevisionMultiInputs to TensorSchema (#21844) · a70d0bd0
  Benji Beck authored Aug 19, 2025
```
Signed-off-by: Benji Beck <benjibeck@meta.com>
```
  a70d0bd0
- [Model] Removes redundant all-reduce operation in Qwen3MoeSparseMoeBlock (#23169) · 4f510bc2
  yiz-liu authored Aug 20, 2025
```
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
```
  4f510bc2
- [FEAT] [Performance] Enable DP for ViT in Qwen2.5VL (#22742) · 1298c677
  TJian authored Aug 19, 2025
```
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  1298c677
- [Model] support new model ovis2.5 (#23084) · b87cb97a
  myselvess authored Aug 19, 2025
```
Signed-off-by: myselvess <244285088@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  b87cb97a
- [Model] Add multi_label_classification support (#23173) · f856c33c
  wang.yuqi authored Aug 19, 2025
```
Signed-off-by: wang.yuqi <noooop@126.com>
```
  f856c33c
- [bug fix] Fix llama4 spec decoding (#22691) · 5bfe0dea
  qizixi authored Aug 19, 2025
```
Signed-off-by: qizixi <qizixi@meta.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
```
  5bfe0dea
- [Bugfix] Fix broken Minimax-01-VL model (#22116) · 31fd3265
  Isotr0py authored Aug 19, 2025
```
Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  31fd3265
- Fix GLM-4.5V-FP8 numerical issue (#22949) · 4efd43e9
  qizixi authored Aug 19, 2025
```
Signed-off-by: qizixi <qizixi@meta.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  4efd43e9
- [Model] Support Pipeline Parallelism for moonshotai/Kimi-VL-A3B-Thinking-2506 (#23114) · fda9537c
  Jiangyun Zhu authored Aug 19, 2025
```
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  fda9537c
- Migrate InternVLImagePixelInputs (in nemotron_vl.py) to TensorSchema (#22023) · e75f3422
  Benji Beck authored Aug 18, 2025
```
Signed-off-by: Benji Beck <benjibeck@meta.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  e75f3422
- [TPU] make ptxla not imported when using tpu_commons (#23081) · e9d6a3db
  Chengji Yao authored Aug 18, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@gmail.com>
Signed-off-by: Chengji Yao <chengjiyao@google.com>
Co-authored-by: Chengji Yao <chengjiyao@gmail.com>
```
  e9d6a3db
18 Aug, 2025 5 commits
- Use Blackwell FlashInfer MXFP4 MoE by default if available (#23008) · 6d25e3fd
  Michael Goin authored Aug 18, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6d25e3fd
- [Bugfix] Support compile for Transformers multimodal (#23095) · 0e3bb543
  Raushan Turganbay authored Aug 18, 2025
```
Signed-off-by: raushan <raushan@huggingface.co>
```
  0e3bb543
- [Refactor] Get prompt updates earlier (#23097) · d3f71f12
  Cyrus Leung authored Aug 18, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d3f71f12
- [Refactor] Define MultiModalKwargsItems separate from MultiModalKwargs (#23053) · 27e8d1ea
  Cyrus Leung authored Aug 18, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  27e8d1ea
- [Bugfix] fix Qwen2.5-Omni processor output mapping (#23058) · 9f1c6422
  double7 authored Aug 18, 2025
```
Signed-off-by: double7 <33449816+DoubleVII@users.noreply.github.com>
Co-authored-by: 杨森 <yangsen.double7@bytedance.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  9f1c6422
17 Aug, 2025 1 commit
- fix: gptq marlin weight loading failure (#23066) · 0fc8fa75
  Simon Mo authored Aug 17, 2025
  
  0fc8fa75