Commits · 55d27cca55310a04fb82c90d26a5afed90f01de7 · OpenDAS / vllm_cscc · GitLab

09 Mar, 2026 24 commits
- [Misc] fix typo: dependant -> dependent (2 lines change) (#36511) · 55d27cca
  SoluMilken authored Mar 10, 2026
```
Signed-off-by: SoluMilken <ypiheyn.imm02g@g2.nctu.edu.tw>
```
  55d27cca
- [Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA... · 580864d8
  Roberto L. Castro authored Mar 09, 2026
```
[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
```
  580864d8
- [ROCM] Optimize the fused_topk_bias to use aiter instead of fallback torch ops. (#36253) · 70485a11
  Taoyu Zhu authored Mar 10, 2026
```
Signed-off-by: zhutaoyu <zhutaoyu97@gmail.com>
```
  70485a11
- [Bugfix] Clear stale CG keys after memory profiling (#36416) · 00c4cb56
  Matthew Bonanni authored Mar 09, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  00c4cb56
- [Refactor] Simplify `chat_completion_full_generator` for tool parsers (#35634) · 941e52c2
  Wentao Ye authored Mar 09, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  941e52c2
- [Bug] Fix pooling model benchmark script (#36300) · be292b7c
  Wentao Ye authored Mar 09, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  be292b7c
- Reapply [Attention] Refactor `check_and_update_config` (#35122) · 77a73458
  Matthew Bonanni authored Mar 09, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  77a73458
- Support online use_audio_in_video (#36319) · 5578f2a4
  Tianyu Guo authored Mar 09, 2026
```
Signed-off-by: Tianyu Guo <guoty9@mail2.sysu.edu.cn>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  5578f2a4
- [Frontend] Move warmup into Renderer (#36482) · 3ec21150
  Cyrus Leung authored Mar 09, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  3ec21150
- [MM Encoder] Default to use TORCH_SDPA backend for ViT on Volta/Turing GPU (#36472) · b0906d8b
  Isotr0py authored Mar 09, 2026
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  b0906d8b
- [Deprecation][1/2] Remove items deprecated in v0.18 (#36470) · f96c3ab0
  Cyrus Leung authored Mar 09, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  f96c3ab0
- [Kernel] Add fused_sigmoid_gating_delta_rule_update kernel for Qwen3 Next (#35777) · dc6b5784
  Xin Yang authored Mar 08, 2026
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
```
  dc6b5784
- [Frontend] Add Support for MM Encoder/Decoder Beam Search (Online Transcriptions) (#36160) · 65a4da15
  Alex Brooks authored Mar 08, 2026
```
Signed-off-by: Alex Brooks <albrooks@redhat.com>
```
  65a4da15
- [Bugfix] Avoid to replace non-tensor members in cpu model runner (#36430) · 217f2759
  Li, Jiang authored Mar 09, 2026
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  217f2759
- [Frontend][2/n] Improve pooling entrypoints | embed. (#36110) · fff3711a
  wang.yuqi authored Mar 09, 2026
```
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Signed-off-by: wang.yuqi <noooop@126.com>
```
  fff3711a
- [Bugfix] Skip out-of-stage layers in get_layers_from_vllm_config for pipeline parallel (#36243) · c4d859c2
  Tushar Shetty authored Mar 09, 2026
```
Signed-off-by: Tushar Shetty <tushar.shetty@abbyy.com>
Signed-off-by: Tushar Shetty <54362365+tusharshetty61@users.noreply.github.com>
```
  c4d859c2
- feat(attention): extract KV-cache update from FlexAttention backend (#36263) · 74743104
  cong-or authored Mar 09, 2026
```
Signed-off-by: cong-or <conchubhar.gannon@gmail.com>
```
  74743104
- [Misc] Move processors to `transformers_utils` (#35953) · d62856b9
  Cyrus Leung authored Mar 09, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d62856b9
- Increase Flexibility for OOV Multimodal Token Handling (#34858) · bd2659a5
  Alex Brooks authored Mar 08, 2026
```
Signed-off-by: Alex Brooks <albrooks@redhat.com>
```
  bd2659a5
- fix: Use iterator as not to store all the file loads in memory at once (#36149) · 90512b2e
  Shaun Kotek authored Mar 09, 2026
```
Signed-off-by: Shaun Kotek - Nvidia <skotek@nvidia.com>
```
  90512b2e
- [Examples][1/n] Resettle basic examples. (#35579) · dcf8862f
  wang.yuqi authored Mar 09, 2026
```
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Signed-off-by: wang.yuqi <noooop@126.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  dcf8862f
- [Bugfix] Fix CPU OMP autobind assertion to use local_world_size (#35815) · 43aa3892
  Weiguang Li authored Mar 09, 2026
```
Signed-off-by: liweiguang <codingpunk@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
```
  43aa3892
- Allow `markdownlint` to run locally (#36398) · a0f44bb6
  Harry Mellor authored Mar 09, 2026
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  a0f44bb6
- [cudagraph] fix cudagraph warning in deepseekv32 (#28044) · e5ff1402
  Jiangyun Zhu authored Mar 09, 2026
```
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
```
  e5ff1402
08 Mar, 2026 3 commits
- Add support for ModelOpt MXFP8 MoE models (#35986) · 0a6a3a12
  danisereb authored Mar 08, 2026
```
Signed-off-by: Daniel Serebrenik <daserebrenik@nvidia.com>
```
  0a6a3a12
- [Frontend] Add GPU-less render serving path (`vllm launch render`) (#36166) · 4497431d
  Sage authored Mar 08, 2026
  
  4497431d
- [Model] Nano Nemotron VL - fast media preprocessing (#35657) · b7332b05
  nvnbagrov authored Mar 08, 2026
```
Signed-off-by: Natan Bagrov <nbagrov@nvidia.com>
```
  b7332b05
07 Mar, 2026 13 commits
- [LMCache MP Patch]: Race Condition + Duplicated Block Ids (#35831) · 5d6aae45
  Samuel Shen authored Mar 07, 2026
  
  5d6aae45
- [Bugfix][LMCache][KVConnector] fix potential memory leak in LMCache multiprocess mode (#35931) · 63298ee1
  Roy Huang authored Mar 07, 2026
  
  63298ee1
- [compile] Split compile/warmup monitoring (#36098) · 2dde535d
  Richard Zou authored Mar 07, 2026
  
  2dde535d
- [Perf] Support FP8 KV cache for Flashinfer MLA Sparse (#35891) · 379689d5
  Wei Zhao authored Mar 07, 2026
  
  379689d5
- [Core] NGram GPU Implementation compatible with Async Scheduler (#29184) · a6be75db
  PatchyTIS authored Mar 08, 2026
  
  a6be75db
- [UX][Startup] Account for CUDA graphs during memory profiling (#30515) · ebb9cc5f
  Matthew Bonanni authored Mar 07, 2026
  
  ebb9cc5f
- Adding support to Sarvam's MoE models (#33942) · 85f50eb4
  rahul-sarvam authored Mar 08, 2026
```
Signed-off-by: rahul-sarvam <140298821+rahul-sarvam@users.noreply.github.com>
```
  85f50eb4
- [Misc] Remove duplicate parser registration (#36303) · 5261223c
  Taneem Ibrahim authored Mar 07, 2026
```
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
```
  5261223c
- [V0 Deprecation] Remove unused swap_space parameter (#36216) · 00b814ba
  lif authored Mar 07, 2026
```
Signed-off-by: majiayu000 <1835304752@qq.com>
Co-authored-by: mcelrath
```
  00b814ba
- [Bugfix] Fix compressed-tensors quantization failure for DeepSeek-R1 on MI300x (#36247) · ee8a2951
  vllmellm authored Mar 07, 2026
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  ee8a2951
- feat: expose media_io_kwargs at runtime (#34778) · 755356b3
  milesial authored Mar 06, 2026
```
Signed-off-by: Alexandre Milesi <milesial@users.noreply.github.com>
```
  755356b3
- Fix CUDA graph decode capture crash in AITER FlashAttention (#36042) · 1a971808
  Mengtao (Martin) Yuan authored Mar 06, 2026
```
Signed-off-by: Martin Yuan <myuan@meta.com>
Co-authored-by: Martin Yuan <myuan@meta.com>
```
  1a971808
- refine `vllm bench throughput --backend hf` (#35971) · 7eb524e6
  Kunshang Ji authored Mar 07, 2026
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  7eb524e6