Commits · 64deead719cc181a1930982b0a5f4d280c284156 · OpenDAS / vllm_cscc · GitLab

25 Nov, 2025 21 commits
- [Bugfix] [ROCm] [UX]: revert Flex attention backend (#29371) · 64deead7
  vllmellm authored Nov 25, 2025
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  64deead7
- [BugFix] Use unique ids for different transcription prompts (#29372) · 7992324f
  Nick Hill authored Nov 24, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  7992324f
- Display warning only when ROCm version is less than Pytorch required version (#29200) · 40a6f53f
  Inoki authored Nov 25, 2025
```
Signed-off-by: Inoki <inoki@inoki.cc>
```
  40a6f53f
- Fix PoolingParams.skip_reading_prefix_cache type (#29364) · ce58fdc1
  kflu authored Nov 24, 2025
```
Signed-off-by: KFL <kludev@gmail.com>
```
  ce58fdc1
- Add TP CLI argument to multimodal inference examples (#29301) · a21256c4
  Fanli Lin authored Nov 25, 2025
```
Signed-off-by: Lin, Fanli <fanli.lin@intel.com>
```
  a21256c4
- Scheduled removal of `guided_*` config fields (#29326) · 316c8492
  Harry Mellor authored Nov 25, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  316c8492
- [CI/Test Fix] Fix CP tests on Blackwell (#29338) · 2d9ee28c
  Lucas Wilkinson authored Nov 24, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  2d9ee28c
- [Attention] add `_cudagraph_support` for linear attention (#28934) · 81db702e
  Jiangyun Zhu authored Nov 25, 2025
```
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
```
  81db702e
- [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
- [Bugfix] Make deprecated `--task embedding` consistent with `--runner… (#29312) · 87185c88
  Maryam Tahhan authored Nov 25, 2025
```
Signed-off-by: Maryam Tahhan <mtahhan@redhat.com>
```
  87185c88
- [Metrics] Scheduled removal of deprecated metrics (#29330) · 9cf4edae
  Mark McLoughlin authored Nov 25, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  9cf4edae
- [Docker] Optimize Dockerfile: consolidate apt-get and reduce image size by ~200MB (#29060) · 7012d8b4
  汪志鹏 authored Nov 25, 2025
```
Signed-off-by: princepride <wangzhipeng628@gmail.com>
```
  7012d8b4
- [CI][ROCm] Install arctic-inference on ROCm tests (#29344) · 22b42b54
  Divakar Verma authored Nov 24, 2025
```
Signed-off-by: Divakar Verma <divakar.verma@amd.com>
```
  22b42b54
- [ROCm][MLA] enable fp8 MLA decode on ROCm (#28032) · cb7214d8
  gbyu-amd authored Nov 25, 2025
```
Signed-off-by: guanbao <gyu@amd.com>
Signed-off-by: Guanbao Yu <gyu@amd.com>
Signed-off-by: gbyu-amd <Guanbao.Yu@amd.com>
Co-authored-by: guanbao <gyu@amd.com>
```
  cb7214d8
- [Perf][Deepseek] optimize gather_and_maybe_dequant_cache kernel's perf for... · 77e10c9c
  Pleaplusone authored Nov 25, 2025
```
[Perf][Deepseek] optimize gather_and_maybe_dequant_cache kernel's perf for extremely long sequence (#28029)
Signed-off-by: ganyi <ygan@amd.com>
```
  77e10c9c
- [Perf] Disable DeepGEMM MoE by default when TP=8 is used (#29346) · 6f1355a1
  Michael Goin authored Nov 24, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6f1355a1
- Scheduled removal of `ParallelConfig`'s direct child EPLB fields (#29324) · a4ad43ad
  Harry Mellor authored Nov 25, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  a4ad43ad
- [BugFix] Fix duplicate id tool-call race condition (#29355) · a178a0b4
  Nick Hill authored Nov 24, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  a178a0b4
- [XPU] upgrade torch & ipex 2.9 on XPU platform (#29307) · b8328b49
  Kunshang Ji authored Nov 25, 2025
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  b8328b49
- [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
- [UX] Raise error for attn backend of batch invariant (#29348) · 699bca76
  Wentao Ye authored Nov 24, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  699bca76
24 Nov, 2025 19 commits
- [Bugfix] Only use triton_kernels for MXFP4 on SM90 and SM100 (#29339) · c17610e2
  Michael Goin authored Nov 24, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  c17610e2
- [Hybrid Allocator] Better layer padding strategy for gpt-oss eagle (#29303) · 71df2a57
  Chen Zhang authored Nov 24, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  71df2a57
- Remove VLLM_SKIP_WARMUP tip (#29331) · 4dd42db5
  Tyler Michael Smith authored Nov 24, 2025
```
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
```
  4dd42db5
- [Tests] Verify gpt_oss package is installed in harmony tests (#29336) · 84371daf
  Nick Hill authored Nov 24, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  84371daf
- [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
- [XPU]fix Kimi-VL-A3B-thinking on xpu (#29309) · 3cfa63ad
  Yan Ma authored Nov 25, 2025
```
Signed-off-by: Yan Ma <yan.ma@intel.com>
```
  3cfa63ad
- [CI/Build] Moves to cuda-base runtime image while retaining minimal JIT dependencies (#29270) · 4d6afcad
  Benjamin Bartels authored Nov 24, 2025
```
Signed-off-by: bbartels <benjamin@bartels.dev>
Signed-off-by: Benjamin Bartels <benjamin@bartels.dev>
```
  4d6afcad
- [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
- [Multimodal][Qwen3 Omni] Make Qwen3 Omni work with audio-in-video inputs in V1 engine. (#27721) · 839c6b7b
  Chenheli Hua authored Nov 24, 2025
```
Signed-off-by: Chenheli Hua <huachenheli@outlook.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>
```
  839c6b7b
- [MoE][Refactor] Make select_experts a non-static method (#29067) · 8f066146
  bnellnm authored Nov 24, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  8f066146
- [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
- [NIXL] Use config to enable telemetry + NIXL version bump (#29305) · 26a46558
  Nicolò Lucchesi authored Nov 24, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  26a46558
- [Build/CI][DP/EP] Add QWen/Qwen3-30B-A3B-FP8 + EPLB tests to Nightly H100 and B200 (#29195) · e924bbb4
  Varun Sundar Rabindranath authored Nov 24, 2025
```
Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
```
  e924bbb4
- [Bugfix] properly handle nested json with llama3 tool parser (#27701) · 656516c3
  Aydin Abiar authored Nov 24, 2025
```
Signed-off-by: Aydin Abiar <aydin@anyscale.com>
Signed-off-by: Aydin Abiar <62435714+Aydin-ab@users.noreply.github.com>
Co-authored-by: Aydin Abiar <aydin@anyscale.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
```
  656516c3
- [Bugfix] [ROCm] [UX] Reorganize ROCm Backend Selection Logic (#26980) · e48b2e68
  vllmellm authored Nov 24, 2025
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  e48b2e68
- Add option to use unbacked, and backed size obl dynamic shapes for more sounds... · 7a228b53
  Laith Sakka authored Nov 24, 2025
```
Add option to use unbacked, and backed size obl dynamic shapes for more sounds compilation. (#26199)
Signed-off-by: Laith Sakka <lsakka@meta.com>
```
  7a228b53
- Update KServe guide link in documentation (#29258) · f716a153
  Yuan Tang authored Nov 24, 2025
```
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
```
  f716a153
- [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