Commits · 912744d0668405f2e70f5d1de785ad513abf7b13 · OpenDAS / vllm_cscc

"vscode:/vscode.git/clone" did not exist on "7ff65b19003be4955d2d5d1428e7d94d082559d0"

10 Nov, 2025 7 commits

[Fix] optimize visual token mask with caching and multi-token support (#28374) · 912744d0

Ferrebo authored Nov 10, 2025

Signed-off-by: Ferrebo <itachi971009@gmail.com>
Signed-off-by: kebo01 <kebo01@baidu.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

912744d0

[bugfix] fix siglip batch text output error (#28365) · 15be507c
Yu Jiaqi authored Nov 10, 2025
```
Signed-off-by: piood <2477084691@qq.com>
```
15be507c
[Metrics] Refactor LoRA state tracking (#26801) · 6f7de33b
Mark McLoughlin authored Nov 10, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
6f7de33b
Restore PlaMo2 unit test as `pfnet/plamo-2-1b` now supports `transformers >=4.56` (#28019) · a98cc35c
Shinichi Hemmi authored Nov 10, 2025
```
Signed-off-by: Shinichi Hemmi <50256998+Alnusjaponica@users.noreply.github.com>
```
a98cc35c
[V0 deprecation] Remove no longer used `get_metadata_cls` (#28370) · e8697faf
Lucas Wilkinson authored Nov 10, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
e8697faf

[Hardware][AMD][Model] Add Triton MoE tuning support and optimized configs for... · 03fa4d3f

Xiake Sun authored Nov 10, 2025


[Hardware][AMD][Model] Add Triton MoE tuning support and optimized configs for Qwen3 omni for MI308X (#28373)
Signed-off-by: Xiake Sun <xiake.sun@amd.com>
Signed-off-by: Xiake Sun <xisun@amd.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

03fa4d3f

[CI] lora/test_mixtral.py : Add additional expected outputs due to flakiness (#28322) · 6b2b9fd9
Varun Sundar Rabindranath authored Nov 09, 2025
```
Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
```
6b2b9fd9

09 Nov, 2025 12 commits
- [ROCm][Platform] Add RX7900XTX device id in _ROCM_DEVICE_ID_NAME_MAP (#28279) · c5f685b3
  JartX authored Nov 10, 2025
```
Signed-off-by: JartX <sagformas@epdcenter.es>
```
  c5f685b3
- [Kernel] Fix fused_gdn_gating (#28343) · c4768dcf
  Jiangyun Zhu authored Nov 10, 2025
```
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
```
  c4768dcf
- [CI/Build] Temporary fix to LM Eval Small Models (#28324) · a65a934e
  Zhewen Li authored Nov 09, 2025
```
Signed-off-by: zhewenli <zhewenli@meta.com>
```
  a65a934e
- Fix cu_num_generated_tokens slicing logic in LogprobsLists.slice() method (#28214) · 4a8d6bd1
  usberkeley authored Nov 10, 2025
```
Signed-off-by: Bradley <bradley.b.pitt@gmail.com>
```
  4a8d6bd1
- [Core] Separate out attention metadata building logic from prepare inputs (#26764) · 636efd10
  Lucas Wilkinson authored Nov 09, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  636efd10
- [Core] Simplify async KV output aggregation (#28327) · 289eb6c5
  Nick Hill authored Nov 09, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  289eb6c5
- [CI] Fix flaky `test_eagle_correctness` test (#28364) · 19d91ece
  Nicolò Lucchesi authored Nov 09, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  19d91ece
- [Misc] Add some comments in qwen3-next (#28267) · 7ae5a5fb
  Jiangyun Zhu authored Nov 09, 2025
```
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
```
  7ae5a5fb
- [ROCm] Add env to enable/disable aiter triton gemm (#28321) · de2b7830
  Yong Hoon Shin authored Nov 08, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  de2b7830
- [Misc] fix typo and add detailed log (#28178) · e5e9067e
  Ning Xie authored Nov 09, 2025
```
Signed-off-by: Andy Xie <andy.xning@gmail.com>
```
  e5e9067e
- fix: close issue 28338 by fixed python version (#28339) · 3a7d5803
  yihong authored Nov 09, 2025
```
Signed-off-by: yihong0618 <zouzou0208@gmail.com>
```
  3a7d5803
- [chore] Move some wikimedia images to S3 (#28351) · 05f8d690
  Kevin H. Luu authored Nov 08, 2025
```
Signed-off-by: Kevin H. Luu <khluu000@gmail.com>
```
  05f8d690
08 Nov, 2025 21 commits

[Performance][gpt-oss] Revert gpt-oss max cudagraph size to 1024 (#28345) · 404d7a9d
Mohammad Miadh Angkad authored Nov 09, 2025
```
Signed-off-by: Mohammad Miadh Angkad <MAngkad.BSDSBA2027@aim.edu>
```
404d7a9d

[Bugfix] Fix test fused quant layernorm tests (#27865) · 171133f9

ElizaWszola authored Nov 08, 2025


Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

171133f9

Remove setuptools upper bound constraint (<80) (#28337) · 32787d06
Cole Murray authored Nov 08, 2025
```
Signed-off-by: Cole Murray <colemurray.cs@gmail.com>
```
32787d06
[Feat] Drop-in Torch CUDA Profiler (#27841) · 975676d1
Benjamin Chislett authored Nov 08, 2025
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
```
975676d1
Enhance run_cluster.sh for multi-NIC support (#28328) · 77d702a2
Ev Lacey authored Nov 08, 2025
```
Signed-off-by: Ev Lacey <elacey@nvidia.com>
```
77d702a2

[DCP] Support dcp kv_cache interleave size > 1 (#26696) · 2108a571

zhangsicheng5 authored Nov 09, 2025


Signed-off-by: zhangsicheng5 <zhangsicheng5@huawei.com>
Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
Signed-off-by: Qiu <qiuchunshuo@huawei.com>
Co-authored-by: QiuChunshuo <qiuchunshuo@huawei.com>

2108a571

[Bugfix] Spec decode + structured output + spec model max len edge case (#28298) · 47604137
Andy Lo authored Nov 08, 2025
```
Signed-off-by: Andy Lo <andy@mistral.ai>
```
47604137
[Bugfix] Update device name for H200 detection (#28349) · 26990d25
Robert Shaw authored Nov 08, 2025
```
Signed-off-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
```
26990d25
`reasoning_content` -> `reasoning` (#27752) · d9ab1ad9
Harry Mellor authored Nov 08, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
d9ab1ad9
[Attention] Remove max cudagraph size limit of 992 (#27840) · 608bb144
22quinn authored Nov 07, 2025
```
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
```
608bb144
[flashinfer][fix] do not check nvcc availability when using pre-downloaded cubins (#27990) · 4a36681f
Xiaozhu Meng authored Nov 07, 2025
```
Signed-off-by: Xiaozhu <mxz297@gmail.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
```
4a36681f
Refactor CPU/GPU extension targets for CMake build (#28026) · d15afc1f
Abolfazl Shahbazi authored Nov 07, 2025
```
Signed-off-by: Abolfazl Shahbazi <12436063+ashahba@users.noreply.github.com>
```
d15afc1f

[Model] Consolidate Deepseek-MoE implementation with DeepSeek-v2 (#28101) · 934a9c3b

Isotr0py authored Nov 08, 2025


Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>

934a9c3b

[bugfix] support eagle with lora cudagraph specialization (#28318) · 70af44fd
gnovack authored Nov 07, 2025
```
Signed-off-by: gnovack <gnovack@amazon.com>
```
70af44fd

Bump arctic-inference requirement (#28174) · 781f5ebf

Aurick Qiao authored Nov 07, 2025


Co-authored-by: Aurick Qiao <aurick.qiao@snowflake.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

781f5ebf

[Perf][DeepSeek] Add sigmoid+bias fusion to fused_grouped_topk from TRTLLM (#28124) · 08525276

Michael Goin authored Nov 08, 2025

Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

08525276

Update gpu.rocm.inc.md to add support for AMD Ryzen AI MAX / AI 300 Series... · 61d25dc4

Hamid Mukhtar authored Nov 07, 2025


Update gpu.rocm.inc.md to add support for AMD Ryzen AI MAX / AI 300 Series (gfx1151, gfx1150) (#28308)
Signed-off-by: Hamid Mukhtar <15519013+hammmmy@users.noreply.github.com>

61d25dc4

[Bugfix][LoRA][Spec Decode] Support LoRA with speculative decoding (#21068) · d0c77920

Xiaohong (Sean) Chen authored Nov 07, 2025

Signed-off-by: Sean Chen <xiaohong_chen1991@hotmail.com>
Signed-off-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Danielle Robinson <dcmaddix@gmail.com>
Co-authored-by: Haipeng Li <li2haipeng@gmail.com>
Co-authored-by: li2haipeng <44383182+li2haipeng@users.noreply.github.com>

d0c77920

remove resolve_op_overloads and use splitting_ops directly (#28081) · b158df28
Boyuan Feng authored Nov 07, 2025
```
Signed-off-by: Boyuan Feng <boyuan@meta.com>
```
b158df28
[XPU] Enable Expert parallel for MoE models (#28263) · 1aaecda0
Kunshang Ji authored Nov 08, 2025
```
Signed-off-by: Yan Ma <yan.ma@intel.com>
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
1aaecda0
Update Flashinfer from `v0.4.1` to `v0.5.2` (#27952) · 811df41e
Harry Mellor authored Nov 07, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
811df41e