Commits · 56539cddac9eeab0a91941d8de689a6cae5dbe05 · OpenDAS / vllm_cscc

26 Nov, 2025 3 commits
- [Core] Refactor padding logic and pad for CUDA graphs before attention metadata building (#28579) · 56539cdd
  Lucas Wilkinson authored Nov 26, 2025
  
  56539cdd
- [Attention] Remove imports from `vllm/attention/__init__.py` (#29342) · 430dd4d9
  Matthew Bonanni authored Nov 26, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  430dd4d9
- dummy run corner case (#29433) · 12866af7
  Xieyang Xu authored Nov 25, 2025
  
  12866af7
25 Nov, 2025 7 commits

[BugFix] Fix initialization of draft model. (#29319) · de75b0bb

Andrey Khalyavin authored Nov 26, 2025


Signed-off-by: Andrey Khalyavin <halyavin@yandex-team.ru>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>

de75b0bb

[Core] Support logprobs with spec decode + async scheduling (#29223) · 4e57c658
Nick Hill authored Nov 25, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
4e57c658
[Bugfix] Fix overallocation in MM profiling (#29386) · c2c661af
Roger Wang authored Nov 25, 2025
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
c2c661af

[Core] Generalize Encoder-Decoder `seq_lens` computation to avoid Whisper... · 798e87db

Nicolò Lucchesi authored Nov 25, 2025


[Core] Generalize Encoder-Decoder `seq_lens` computation to avoid Whisper hardcoded logic   (#29268)
Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Ekagra Ranjan <3116519+ekagra-ranjan@users.noreply.github.com>

798e87db

EAGLE Support DP>1 (#26086) · 12c007e2

Rémi Delacourt authored Nov 25, 2025


Signed-off-by: Rémi Delacourt <remi@mistral.ai>
Signed-off-by: Rémi Delacourt <54138269+Flechman@users.noreply.github.com>
Signed-off-by: remi <remi@mistral.ai>

12c007e2

[Model] Add HunyuanOCR support (#29327) · 92effb07

Isotr0py authored Nov 25, 2025


Signed-off-by: manayang <jackmanayang@gmail.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: sergeywang <sergeywang@tencent.com>
Co-authored-by: manayang <jackmanayang@gmail.com>
Co-authored-by: manayang <manayang@tencent.com>
Co-authored-by: Roger Wang <hey@rogerw.io>

92effb07

[Spec Decode] Add support for EAGLE3 heads that do not use_aux_hidden_states (#27688) · 5f9679a4

Hanjie Qiu authored Nov 24, 2025


Signed-off-by: hjjq <hanjieq@nvidia.com>
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Co-authored-by: Benjamin Chislett <bchislett@nvidia.com>

5f9679a4

24 Nov, 2025 6 commits
- [Model Runner V2] Simplify Eagle bookkeeping with num_rejected (#29347) · f32c7d6f
  Woosuk Kwon authored Nov 24, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  f32c7d6f
- [Model Runner V2] Add minor clarification comments for Eagle (#29332) · 97588c4d
  Woosuk Kwon authored Nov 24, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  97588c4d
- [Model Runner V2] Change Numba AoT to JIT (#29328) · cec418b5
  Woosuk Kwon authored Nov 24, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  cec418b5
- [Model Runner V2] Implement Single-step Eagle 1 (#29300) · cc313cb7
  Woosuk Kwon authored Nov 24, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  cc313cb7
- [EPLB] Optimize EPLB for Async Rearrange Experts (#22179) · 2601f18a
  WeiQing Chen authored Nov 24, 2025
```
Signed-off-by: David Chen <530634352@qq.com>
Co-authored-by: SunChenxiang123 <1291824390@qq.com>
```
  2601f18a
- [Doc]: fix typos in various files (#29230) · eca7a8fb
  Didier Durand authored Nov 24, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  eca7a8fb
23 Nov, 2025 5 commits
- [Model Runner V2] Add apply_temperature option to gumbel_sample (#29276) · 3e1ad406
  Woosuk Kwon authored Nov 23, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  3e1ad406
- [Model Runner V2] Optimize CUDA graph capture time (#29275) · 62d54ba4
  Woosuk Kwon authored Nov 23, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  62d54ba4
- [Model Runner V2] Support spec decoding [1/N] (#29274) · b004c004
  Woosuk Kwon authored Nov 23, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  b004c004
- [Model Runner V2] Change bookkeeping logic in preparation for spec decoding (#29194) · 7f12c82f
  Woosuk Kwon authored Nov 23, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  7f12c82f
- [Model Runner V2] Minor fix for cudagraph_utils (#29256) · 20ee418a
  Woosuk Kwon authored Nov 22, 2025
  
  20ee418a
22 Nov, 2025 7 commits
- Refactor: Move CUDA graph dispatch logic earlier (#27382) · df78aeef
  Yizhou authored Nov 23, 2025
```
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
```
  df78aeef
- [BugFix] Fix chunked prompt logprobs + preemption (#29071) · 7df331c6
  Nick Hill authored Nov 22, 2025
  
  7df331c6
- [BugFix] Fix returned logprobs with spec decode + prefill chunking (#29216) · d44a63c6
  Nick Hill authored Nov 22, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  d44a63c6
- [Attention] Refactor FA `block_size` limitations to hybrid models only (#29084) · 066209a0
  Nicolò Lucchesi authored Nov 22, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  066209a0
- [Model Runner V2] Limit cudagraph size to max decode batch size (#29221) · e9056056
  Woosuk Kwon authored Nov 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  e9056056
- [Bugfix] Fix block size in block_table with PCP (#29094) · 5c8f2adf
  Jie Luo authored Nov 22, 2025
```
Signed-off-by: Livinfly <luojie3m@gmail.com>
```
  5c8f2adf
- [CI] Bug: Fix triton import issue (#29202) · 1d34eb11
  Wentao Ye authored Nov 21, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  1d34eb11
21 Nov, 2025 8 commits
- [Model Runner V2] Optimize Gumbel Sampling Kernel (#29210) · e9af6ba6
  Woosuk Kwon authored Nov 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  e9af6ba6
- [Chore] Fix pre-commit error after #25266 (#29190) · 1bed891f
  Woosuk Kwon authored Nov 21, 2025
  
  1bed891f
- GPU Model Runner V2 (#25266) · 30b44a15
  Woosuk Kwon authored Nov 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  30b44a15
- [Misc] Move dynamic seed initialization to `EngineArgs` (#29165) · d7219bcd
  Cyrus Leung authored Nov 21, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d7219bcd
- [Core] Align whisper closer to other multimodal models (#27292) · cca2d2cd
  Russell Bryant authored Nov 21, 2025
```
Signed-off-by: Russell Bryant <rbryant@redhat.com>
```
  cca2d2cd
- Revert "[Redo] #26368 (#28771)" (#29121) · 30b9c677
  Jialin Ouyang authored Nov 20, 2025
```
Signed-off-by: Jialin Ouyang <Jialin.Ouyang@gmail.com>
```
  30b9c677
- [CI] Fix mypy for `vllm/v1/worker` (#29037) · 56669c1f
  Wentao Ye authored Nov 20, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  56669c1f
- [LoRA][2/2]Remove LoRA extra vocab (#28545) · 9875be64
  Jee Jee Li authored Nov 21, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  9875be64
20 Nov, 2025 4 commits
- [KVConnector][Core] Support cross-layer KV blocks (#27743) · 64746471
  Or Ozeri authored Nov 20, 2025
```
Signed-off-by: Or Ozeri <oro@il.ibm.com>
```
  64746471
- [ROCm] Add AMD GPU support on Deepseek v3.2 and SparseMLA (#26670) · 06c20c99
  Pleaplusone authored Nov 20, 2025
```
Signed-off-by: ganyi <ygan@amd.com>
```
  06c20c99
- [Bugfix] Revert custom attention mask for gemma3-mm (#28995) · 64192d56
  Isotr0py authored Nov 20, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  64192d56
- [Feat] Iteration-level profiling for Torch and CUDA profiler (#28987) · fcbcba6c
  Benjamin Chislett authored Nov 19, 2025
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  fcbcba6c