Commits · e261d37c9a5e88a6c86d32decf39f1fab7ca1f2c · OpenDAS / vllm_cscc · GitLab

05 Nov, 2025 8 commits
- [Refactor] Lazy-loaded reasoning_parser (#28092) · e261d37c
  Chauncey authored Nov 05, 2025
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
  e261d37c
- [Model, Core] Support Granite Speech & LoRA for STT (#24455) · b7cbc254
  Alex Brooks authored Nov 05, 2025
  
  b7cbc254
- [BugFix] Fix DCP Assert (AssertionError: DCP not support reorder_batch_threshold > 1 now.) (#28100) · d43ad5a7
  Lucas Wilkinson authored Nov 05, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  d43ad5a7
- [Bugfix] Fix encoder-only model support for transformers backend (#28021) · 0ff05e37
  Isotr0py authored Nov 05, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  0ff05e37
- [V0 deprecation] Remove VLLM_USE_V1 usage in most modules (#27955) · 428bc7bf
  wangxiyuan authored Nov 05, 2025
```
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
```
  428bc7bf
- [CI/Build] Enable some fixed tests in AMD CI (#28078) · 878fd5a1
  Zhewen Li authored Nov 04, 2025
```
Signed-off-by: zhewenli <zhewenli@meta.com>
```
  878fd5a1
- [XPU] Add gpt-oss model support for Intel GPU (#27786) · 18b39828
  Kunshang Ji authored Nov 05, 2025
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  18b39828
- [Qwen3-Next] MOE configs for A100-SXM4-80GB TP4 TP8 (#27740) · 4ea62b77
  tou authored Nov 05, 2025
  
  4ea62b77
04 Nov, 2025 31 commits
- Revert "[PERF] Decouple projections from GDN custom op" (#28080) · d4e547bb
  Vadim Gimpelson authored Nov 05, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
```
  d4e547bb
- [ROCm] gemm_a16w16 upstreaming (#26969) · 2d977a7a
  Aleksandr Malyshev authored Nov 04, 2025
```
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
```
  2d977a7a
- [Multimodal] Make MediaConnector extensible. (#27759) · 1fb4217a
  Chenheli Hua authored Nov 04, 2025
```
Signed-off-by: Chenheli Hua <huachenheli@outlook.com>
```
  1fb4217a
- Added disable rule to track files under benchmarks/lib (#28048) · 611c86ea
  nadavkluger authored Nov 04, 2025
```
Signed-off-by: Nadav Kluger <nadav.k@fmr.ai>
```
  611c86ea
- [ROCm][Perf] New design on ROCm AITER MHA backend Implementation (#25763) · dc937175
  Pleaplusone authored Nov 05, 2025
```
Signed-off-by: ganyi <ygan@amd.com>
```
  dc937175
- Remove deprecated `--rope-scaling` and `--rope-theta` (#28006) · 2f1cc8ce
  Harry Mellor authored Nov 04, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  2f1cc8ce
- [AsyncScheduling] Don't schedule past request max_tokens (#27922) · 938a8169
  Nick Hill authored Nov 04, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  938a8169
- [PerfFix] Avoid separate thread for MP executor shm spin (#28012) · c9f66da8
  Nick Hill authored Nov 04, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  c9f66da8
- [model] Add support for openPangu_Ultra_MoE (#27521) · 05cae69f
  yt0428 authored Nov 05, 2025
```
Signed-off-by: yuantao <2422264527@qq.com>
Signed-off-by: yt0428 <51468697+yt0428@users.noreply.github.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  05cae69f
- [PERF] Decouple projections from GDN custom op (#27512) · 5fd8f02e
  Vadim Gimpelson authored Nov 04, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
```
  5fd8f02e
- [Perf] SM100 - add swap AB optimization to CUTLASS FP8 GEMM (#27284) · 97e3dda8
  lyrisz authored Nov 04, 2025
```
Signed-off-by: Faqin Zhong <faqin.zhong@gmail.com>
Co-authored-by: Faqin Zhong <zhofaqin@amazon.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  97e3dda8
- [BugFix] Fix incorrect preallocated sampled_token_ids tensor size (#28025) · 5a0a6dfd
  Nick Hill authored Nov 04, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  5a0a6dfd
- [Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123) · 938772af
  bnellnm authored Nov 04, 2025
  
  938772af
- [Model] add optimal triton fused moe configs for NemotronH MoE (#27967) · e4ee6586
  tomeras91 authored Nov 04, 2025
```
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
```
  e4ee6586
- [Model][Bugfix] fix pipeline parallelism support for NemotronH (#27968) · 77f8001f
  tomeras91 authored Nov 04, 2025
```
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
```
  77f8001f
- [Core] Enable StatLogger in LLMEngine (#28020) · 300a2659
  Zhuohan Li authored Nov 04, 2025
```
Signed-off-by: Zhuohan Li <zhuohan123@gmail.com>
```
  300a2659
- Support using Int4PreshuffledTensor after loading (#26066) · 03c4c4aa
  Jerry Zhang authored Nov 04, 2025
```
Signed-off-by: Jerry Zhang <jerryzh168@gmail.com>
```
  03c4c4aa
- Load tuned fused_moe_lora shrink and expand kernel configs separately (#27435) · 2ec401bc
  yugong333 authored Nov 04, 2025
```
Signed-off-by: Yu Gong <yu3.gong@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  2ec401bc
- [BugFix][Performance] Restore flashinfer autotuning for all scenarios (#27904) · 4022a9d2
  Varun Sundar Rabindranath authored Nov 04, 2025
  
  4022a9d2
- [CI/Build] Fix OpenAI API correctness on AMD CI (#28022) · 53f6e81d
  Zhewen Li authored Nov 03, 2025
```
Signed-off-by: zhewenli <zhewenli@meta.com>
```
  53f6e81d
- [Model] fix ernie45 reasoning_parser (#27973) · 43a6acfb
  CSWYF3634076 authored Nov 04, 2025
```
Signed-off-by: wangyafeng <wangyafeng@baidu.com>
```
  43a6acfb
- [KV Connector] Make KVCacheConfig an explicit constructor argument (#27887) · 58279c60
  Mark McLoughlin authored Nov 04, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  58279c60
- [CI/Build] Update LM Eval Version in AMD CI (#27944) · 2f84ae1f
  Zhewen Li authored Nov 03, 2025
```
Signed-off-by: zhewenli <zhewenli@meta.com>
```
  2f84ae1f
- [CPU]Improve dynamic 4bit moe performance (#27240) · f32cbc9a
  xiangze-arm authored Nov 04, 2025
```
Signed-off-by: Zhang Xiangze <Xiangze.Zhang@arm.com>
```
  f32cbc9a
- [Bug] Batch invariant: Fix flash attn MLA `RuntimeError: scheduler_metadata... · 7e4be741
  Wentao Ye authored Nov 04, 2025
```
[Bug] Batch invariant: Fix flash attn MLA `RuntimeError: scheduler_metadata must have shape (metadata_size)` (#27884)
```
  7e4be741
- [Metrics] Enable sleep state metric outside of dev mode (#27867) · 380ba681
  Mark McLoughlin authored Nov 04, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  380ba681
- [NIXL][XPU] Pin NIXL version to 0.7.0 (#27849) · 14a125a0
  liuzhenwei authored Nov 04, 2025
```
Signed-off-by: zhenwei-intel <zhenwei.liu@intel.com>
```
  14a125a0
- [Refactor] Lazy import tool_parser (#27974) · c02fccdb
  Chauncey authored Nov 04, 2025
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
  c02fccdb
- [LoRA] Lora shrink swizzle (#27694) · 6ddae740
  li2haipeng authored Nov 03, 2025
```
Signed-off-by: li2haipeng <44383182+li2haipeng@users.noreply.github.com>
Signed-off-by: Haipeng Li <li2haipeng@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  6ddae740
- [Bugfix][ROCm] Fix ViT rotary embeddings for torch.compile compatibility on ROCm (#27748) · b13a4475
  vllmellm authored Nov 04, 2025
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  b13a4475
- Remove the tpu docker image nightly build. (#27997) · 7956b0c0
  QiliangCui authored Nov 03, 2025
```
Signed-off-by: Qiliang Cui <derrhein@gmail.com>
```
  7956b0c0
03 Nov, 2025 1 commit
- [Bugfix] Fix MoE Routing Simulation (#28002) · 37587573
  Tyler Michael Smith authored Nov 03, 2025
```
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
```
  37587573