Commits · dc2979c58574e7a49d17b50c5770010039145aac · OpenDAS / vllm_cscc · GitLab

18 Sep, 2025 10 commits

[Kernels] Overlap shared experts with combine instead of dispatch (#24254) · dc2979c5
bnellnm authored Sep 18, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
dc2979c5

[Bugfix][Qwen3-Next] add prefixes to shared_expert in qwen3-next and mlp in... · 027d37df

toncao authored Sep 18, 2025


[Bugfix][Qwen3-Next] add prefixes to shared_expert in qwen3-next and mlp in qwen2moe to successfully load ignored params in quantized models (#24960)
Signed-off-by: toncao <cpatonn@gmail.com>
Co-authored-by: toncao <cpatonn@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>

027d37df

[Core][MM] Cleanup `MultiModalCache` (#25006) · b9821967
Lukas Geiger authored Sep 18, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
b9821967
[Docs] Clean up the contributing README (#25099) · 32baf1d0
Harry Mellor authored Sep 18, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
32baf1d0

[MM Encoder] Apply DP ViT for Qwen3-VL model series (#24955) · 3127274d

Roger Wang authored Sep 17, 2025

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

3127274d

[Kernels] Enable DeepGEMM by default (#24462) · 4ac510f4
bnellnm authored Sep 17, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
4ac510f4
[V0 Deprecation] Skip PP test (#25128) · 7fb2a5be
Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
7fb2a5be
[V0 Deprecation] Remove misc V0 tests (#25118) · 6c036615
Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
6c036615
[V0 Deprecation] Remove V0 Tracing & Metrics tests (#25115) · 2fc24e94
Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
2fc24e94
[V0 Deprecation] Remove V0 Engine tests (#25114) · 2c3c1bd0
Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
2c3c1bd0

17 Sep, 2025 30 commits
- [Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537) · 5963b98b
  bnellnm authored Sep 17, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  5963b98b
- [Bugfix] Fix accuracy issue for silu_mul + nvfp4 quant fusion kernel (#24833) · e6585ddb
  elvischenv authored Sep 18, 2025
```
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
```
  e6585ddb
- Add a batched auto tune script (#25076) · 2a4d6412
  Karan Goel authored Sep 17, 2025
```
Signed-off-by: Karan Goel <karangoel@google.com>
Signed-off-by: Karan Goel <3261985+karan@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  2a4d6412
- [Bugfix] Refactor Flashinfer TRTLLM attention kernel selection logic (#24600) · e67a79db
  elvischenv authored Sep 18, 2025
```
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  e67a79db
- Disable failing GPT-OSS Eval (Blackwell) for now (#25107) · 9f882d87
  Michael Goin authored Sep 17, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  9f882d87
- Aiter mha fp8 fix (#24991) · 1a456c7c
  Douglas Lehr authored Sep 17, 2025
```
Signed-off-by: Doug Lehr <douglehr@amd.com>
Co-authored-by: Doug Lehr <douglehr@amd.com>
```
  1a456c7c
- [Bugfix][B200] Fix `cutlass_mla` hang (#24966) · fedb75fa
  Alexander Matveev authored Sep 17, 2025
```
Signed-off-by: Alexander Matveev <amatveev@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  fedb75fa
- [gpt-oss][2] fix types for streaming (#24556) · bff2e5f1
  Andrew Xia authored Sep 17, 2025
```
Signed-off-by: Andrew Xia <axia@meta.com>
```
  bff2e5f1
- [Kernel] Faster pre-processing time for W4A8 (#23972) · 3c068c63
  czhu-cohere authored Sep 17, 2025
```
Signed-off-by: czhu-cohere <conway.zhu@cohere.com>
```
  3c068c63
- [BUG] Exclude .pth files when pulling remote files (#25092) · f20c3b09
  ahao-anyscale authored Sep 17, 2025
```
Signed-off-by: ahao-anyscale <ahao@anyscale.com>
```
  f20c3b09
- [Bugfix] Update import path for bc_linter_include (#24766) · 88313154
  Mohammad Miadh Angkad authored Sep 18, 2025
```
Signed-off-by: Mohammad Miadh Angkad <mangkad.bsdsba2027@aim.edu>
```
  88313154
- [Misc] Update owners for KV connector and V1 offloading (#25041) · ee5fd491
  Yihua Cheng authored Sep 17, 2025
```
Signed-off-by: ApostaC <yihua98@uchicago.edu>
```
  ee5fd491
- [V1] Logits processor docs (#22919) · 7ae98875
  afeldman-nm authored Sep 17, 2025
```
Signed-off-by: Andrew Feldman <afeldman@redhat.com>
Signed-off-by: afeldman-nm <156691304+afeldman-nm@users.noreply.github.com>
Co-authored-by: Joseph Marinier <Joseph.Marinier@gmail.com>
```
  7ae98875
- [CI Bugfix] Fix failing test_model_load_with_params tests due to tokenizer refactor (#25086) · e3db5ebb
  Michael Goin authored Sep 17, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  e3db5ebb
- [V0 Deprecation] Remove V0 tests in test_sequence.py (#25088) · 9d442b7c
  Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  9d442b7c
- [CI] Revert back prepare_prompts and check_answers (#25087) · eb68c2dc
  Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  eb68c2dc
- Change log level from info to debug for IOProcessor (#24999) · 8b32464a
  Michael Goin authored Sep 17, 2025
```
Signed-off-by: Michael Goin <mgoin64@gmail.com>
```
  8b32464a
- [V0 Deprecation] Remove unused output processor util (#25023) · 99cc41ad
  Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
```
  99cc41ad
- Remove unused find_cuda_init helper script (#25044) · d6a518fd
  Simon Mo authored Sep 17, 2025
  
  d6a518fd
- cleanup: remove adapter commons (#25045) · 4aa8c7b0
  Simon Mo authored Sep 17, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  4aa8c7b0
- [V0 Deprecation] Remove V0 Core tests (#25082) · 4b946d69
  Woosuk Kwon authored Sep 17, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  4b946d69
- [CI Bugfix] Fix failing test_invalid_env (#25078) · 087c6ffc
  Michael Goin authored Sep 17, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  087c6ffc
- [Docs] vllm/benchmarks/datasets.py fix docstring param format. (#24970) · 4a2d33e3
  samzong authored Sep 17, 2025
```
Signed-off-by: samzong <samzong.lu@gmail.com>
```
  4a2d33e3
- Remove old cutlass mla (#23961) · 8f3616f4
  Matthew Bonanni authored Sep 17, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  8f3616f4
- [Docs] improve code formatting and comments for eliminate griffe build warning. (#25010) · 47f670b0
  samzong authored Sep 17, 2025
```
Signed-off-by: samzong <samzong.lu@gmail.com>
```
  47f670b0
- [Bugfix][Qwen3-Next] fixes the varlen issue in qwen3-next's MTP implementation. (#24957) · dd6a910a
  Tao He authored Sep 17, 2025
```
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
```
  dd6a910a
- [fix] lora benchmarks pass no_lora_flag_cpu (#23774) · 1b962e24
  dolpm authored Sep 17, 2025
```
Signed-off-by: Dylan Maloy <34420038+dolpm@users.noreply.github.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  1b962e24
- Apply fixes for CUDA 13 (#24599) · bfe93801
  Aidyn-A authored Sep 17, 2025
```
Signed-off-by: Aidyn-A <aidyn.b.aitzhan@gmail.com>
```
  bfe93801
- [Bugfix] Fix Stream usage in CPU model runner and OneDNN kernel check (#25046) · 9fccd04e
  Li, Jiang authored Sep 17, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  9fccd04e
- Add RADIO Vision Encoder Support to vLLM (#24595) · 252ada55
  danielafrimi authored Sep 17, 2025
```
Signed-off-by: Daniel Afrimi <danielafrimi8@gmail.com>
Co-authored-by: root <root@cw-dfw-h100-001-305-026.cm.cluster>
```
  252ada55