Commits · 5576227bc132996d9e4ca279a8a8cf9e7b8a8ca4 · OpenDAS / vllm_cscc · GitLab

08 Jan, 2026 21 commits
- [Model] Standardize common vision encoders (#31947) · 5576227b
  Cyrus Leung authored Jan 08, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  5576227b
- [Chore] Further cleanup pooler (#31951) · d1b6fe00
  Cyrus Leung authored Jan 08, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d1b6fe00
- RayLLM Bugfix - Preserve obj store URL for multi engine_config creation (#30803) · 04a49669
  omer-dayan authored Jan 08, 2026
```
Signed-off-by: Omer Dayan <omdayan@nvidia.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  04a49669
- [Misc] Support qwen3-next lora (#31719) · 96fcd3c2
  BingjiaWang authored Jan 08, 2026
  
  96fcd3c2
- fix(compile): apply partition wrapper when loading AOT cached functions (#31536) · 1f214290
  DevByteAI authored Jan 08, 2026
```
Signed-off-by: Devbyteai <abud6673@gmail.com>
Signed-off-by: DevByteAI <161969603+devbyteai@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
```
  1f214290
- Decouple page_size_bytes calculation in AttentionSpec for TPU/RPA Compatibility. (#31635) · b634e619
  Lumosis authored Jan 08, 2026
```
Signed-off-by: Lihao Ran <imlihao.ran@gmail.com>
Signed-off-by: Lumosis <30372757+Lumosis@users.noreply.github.com>
```
  b634e619
- [Models] Allow converting Qwen3-VL into Reranker model (#31890) · eac3b96e
  Isotr0py authored Jan 08, 2026
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  eac3b96e
- fix(rocm): Add get_supported_kernel_block_sizes() to ROCM_ATTN (#31712) · 107cf8e9
  Rabi Mishra authored Jan 08, 2026
```
Signed-off-by: rabi <ramishra@redhat.com>
```
  107cf8e9
- [Model] Enable LoRA support for tower and connector in GLM4-V (#31652) · 63baa28c
  Zyyeric authored Jan 08, 2026
```
Signed-off-by: Zyyeric <eric1976808123@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  63baa28c
- [Bugfix] Remove the num_hidden_layers override for glm4_moe (#31745) · e5173d3b
  Andy Liu authored Jan 07, 2026
  
  e5173d3b
- [Fix] Enable mm_processor_cache with vision LoRA (#31927) · d3235cb5
  prashanth058 authored Jan 08, 2026
```
Signed-off-by: prashanth058 <prashanth.dannamaneni@uipath.com>
```
  d3235cb5
- [BugFix] Fix spec decoding edge case bugs (#31944) · 287b37cd
  Nick Hill authored Jan 07, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  287b37cd
- [grpc] Support gRPC server entrypoint (#30190) · 791b2fc3
  Chang Su authored Jan 07, 2026
```
Signed-off-by: Chang Su <chang.s.su@oracle.com>
Signed-off-by: njhill <nickhill123@gmail.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: njhill <nickhill123@gmail.com>
Co-authored-by: Simon Mo <simon.mo@hey.com>
```
  791b2fc3
- [platform] add dp_metadata arg to set_additional_forward_context (#31942) · 2ab441be
  Ronald authored Jan 08, 2026
```
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
```
  2ab441be
- [Model] Enable LoRA support for tower and connector in DotsOCR (#31825) · 9572f74f
  ShaanveerS authored Jan 08, 2026
```
Signed-off-by: ShaanveerS <shaanver.singh@gmail.com>
```
  9572f74f
- [ROCm][LoRA] Fix MoE accuracy regression by preserving float32 router weight scaling (#31931) · c4041f37
  Andreas Karatzas authored Jan 07, 2026
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  c4041f37
- [MoE Refactor][16/N] Apply Refactor to NVFP4 (#31692) · 9f6dcb71
  Robert Shaw authored Jan 07, 2026
```
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Signed-off-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Pavani Majety <pmajety@nvidia.com>
```
  9f6dcb71
- fix(rocm): add early return in get_flash_attn_version for ROCm (#31286) · 39d82005
  Rabi Mishra authored Jan 08, 2026
```
Signed-off-by: rabi <ramishra@redhat.com>
```
  39d82005
- feat(moe): Add is_act_and_mul=False support for Triton MoE kernels (#31645) · 25eef3dc
  Rabi Mishra authored Jan 08, 2026
```
Signed-off-by: rabi <ramishra@redhat.com>
```
  25eef3dc
- [0/N][Attention] Fix miscellaneous pre-commit issues (#31924) · 0d766741
  Matthew Bonanni authored Jan 07, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  0d766741
- [MoE Refactor][15/N] Apply Refactor to Fp8 (#31415) · 5dcd7ef1
  Robert Shaw authored Jan 07, 2026
  
  5dcd7ef1
07 Jan, 2026 19 commits
- Add back missing DeepEP LL params (#31911) · ffc0a279
  Elvir Crnčević authored Jan 07, 2026
```
Signed-off-by: Elvir Crncevic <elvircrn@gmail.com>
```
  ffc0a279
- [BugFix] Fix bad words with speculative decoding (#31908) · 10ef65ed
  Nick Hill authored Jan 07, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  10ef65ed
- [EPLB] Optimize EPLB with numpy (#29499) · 6170d47d
  Ilya Markov authored Jan 07, 2026
```
Signed-off-by: ilmarkov <markovilya197@gmail.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  6170d47d
- [Kernel] Support bias type in grouped_topk kernel (#31781) · 0ada960a
  Xin Yang authored Jan 07, 2026
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  0ada960a
- [refactor] refactor memory constants usage (#31865) · c907d221
  Ning Xie authored Jan 08, 2026
```
Signed-off-by: Andy Xie <andy.xning@gmail.com>
```
  c907d221
- Enable quantized attention in NemotronH models (#31898) · bf184a66
  roikoren755 authored Jan 07, 2026
```
Signed-off-by: Roi Koren <roik@nvidia.com>
```
  bf184a66
- UX: add vLLM env info in '/server_info' (#31899) · 30399cc7
  Jee Jee Li authored Jan 08, 2026
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  30399cc7
- [KVConnector]: Enable Cross-layers KV cache layout for MultiConnector (#30761) · b89443b8
  Kfir Toledo authored Jan 07, 2026
```
Signed-off-by: Kfir Toledo <kfir.toledo@ibm.com>
```
  b89443b8
- [Bugfix]: prevent leaking tokens in crash log (#30751) · 1d9e9ae8
  Marko Rosenmueller authored Jan 07, 2026
```
Signed-off-by: Marko Rosenmueller <5467316+dr75@users.noreply.github.com>
```
  1d9e9ae8
- [Refactor] Clean up pooler modules (#31897) · b7036c87
  Cyrus Leung authored Jan 08, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  b7036c87
- [Perf][Kernels] Enable FlashInfer DeepGEMM swapAB on SM90 (for W8A8 Linear Op) (#29213) · cc6dafae
  Kate Cheng authored Jan 07, 2026
```
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
Signed-off-by: Jhao-Ting Chen <jhaotingc@nvidia.com>
Co-authored-by: Jhao-Ting Chen <jhaotingc@nvidia.com>
```
  cc6dafae
- [OpenAI] Extend VLLMValidationError to additional validation parameters (#31870) · 1ab055ef
  R3hankhan authored Jan 07, 2026
```
Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>
```
  1ab055ef
- [Chore] Migrate V0 attention utils (#31891) · b665bbc2
  Cyrus Leung authored Jan 07, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  b665bbc2
- [Refactor] GLM-ASR Modeling (#31779) · 97413875
  Jared Wen authored Jan 07, 2026
```
Signed-off-by: JaredforReal <w13431838023@gmail.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  97413875
- [ROCm][AITER] fix wrong argument passed to AITER `flash_attn_varlen_func` (#31880) · 41cfa506
  vllmellm authored Jan 07, 2026
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  41cfa506
- [Bugfix][MTP] Fix GLM4 MoE fp8 loading with MTP on (#31757) · d111bc53
  Andy Liu authored Jan 07, 2026
```
Signed-off-by: Andy Liu <andyliu@roblox.com>
```
  d111bc53
- [Misc] Improve error messages for unsupported types and parameters (#30593) · 0790f076
  BlankR authored Jan 07, 2026
```
Signed-off-by: BlankR <hjyblanche@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
```
  0790f076
- [Model] Cleanup: Remove redundant manual definition of... · 1f33e38e
  maang authored Jan 07, 2026
```
[Model] Cleanup: Remove redundant manual definition of `make_empty_intermediate_tensors` in GLM-4-MoE (#31869)
Signed-off-by: maang <maang_h@163.com>
```
  1f33e38e
- [XPU]fallback to TRITON_ATTN on xpu when use float32 dtype (#31762) · 59fe6f29
  sihao_li authored Jan 07, 2026
```
Signed-off-by: sihao.li <sihao.li@intel.com>
```
  59fe6f29