Commits · 84166fee9770e6fba71a96978b3e7d149392fb28 · OpenDAS / vllm_cscc

07 Jun, 2025 1 commit

[Kernel] Integrate CUTLASS MoE kernel with PPLX (#18762) · 84166fee

ElizaWszola authored Jun 07, 2025


Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

84166fee

06 Jun, 2025 17 commits
- [BugFix] Fix tpu_model_runner block_id concatenation (#19228) · 46ecc579
  Nick Hill authored Jun 06, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  46ecc579
- [Core] Fix abrupt request abort (#18485) · b6a3a9f7
  Nicolò Lucchesi authored Jun 07, 2025
```
Signed-off-by: nicklucche <nlucches@redhat.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  b6a3a9f7
- [Bugfix][Core] Update cancellation logic in `generate()` to handle Generator exits (#19225) · ca27f0f9
  Adolfo Victoria authored Jun 06, 2025
```
Co-authored-by: Adolfo Victoria <adovi@meta.com>
```
  ca27f0f9
- [BugFix] Fix MultiConnector test after HMA changes (#19291) · aad30bd3
  Nick Hill authored Jun 06, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  aad30bd3
- improve logits bias (#19041) · 8267f991
  Yu Guo authored Jun 06, 2025
  
  8267f991
- [Core] Raise when non-multi-instance DP clients target a DP rank (#19227) · 7353492a
  jmswen authored Jun 06, 2025
```
Signed-off-by: Jon Swenson <jmswen@gmail.com>
```
  7353492a
- [Model] Optimize nemotron_h implementation (#19249) · 7661e92e
  Jee Jee Li authored Jun 06, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  7661e92e
- Fix CompilationConfig repr (#19091) · da511d54
  Richard Zou authored Jun 06, 2025
```
Signed-off-by: rzou <zou3519@gmail.com>
```
  da511d54
- [Docs] Improve V1 KVConnector interface documentation (#19172) · 65c69444
  Nick Hill authored Jun 06, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  65c69444
- [Quantization] Bump compressed-tensors version; update NVFP4A16 test model (#19224) · 94870359
  Dipika Sikka authored Jun 06, 2025
```
Signed-off-by: Dipika Sikka <dipikasikka1@gmail.com>
```
  94870359
- [TPU] fix kv cache dtype in model runner (#19244) · 0d49483e
  Chengji Yao authored Jun 06, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@google.com>
```
  0d49483e
- [v1][P/D] Fix a edge case in kv cache schedule (#19182) · 90b78ec5
  Jinghui Zhang authored Jun 05, 2025
```
Co-authored-by: jinghui <jinghui@fb.com>
```
  90b78ec5
- Support allowed_token_ids in ChatCompletionRequest (#19143) · 3da2313d
  Xu Song authored Jun 06, 2025
```
Signed-off-by: Xu Song <xusong.vip@gmail.com>
```
  3da2313d
- [TPU] update torch_xla pin (#19231) · b61dc5f9
  Chengji Yao authored Jun 05, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@google.com>
```
  b61dc5f9
- [v1] Hybrid Memory Allocator (#17996) · f8a1a2d1
  Chen Zhang authored Jun 06, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  f8a1a2d1
- [Bugfix] Fix EAGLE vocab embedding construction for Llama 70B (#19033) · 3465b87e
  Benjamin Chislett authored Jun 05, 2025
```
Signed-off-by: Benjamin Chislett <benjamin.chislett@centml.ai>
```
  3465b87e
- Fix AOPerModuleConfig name changes (#18869) · c8134bea
  Jerry Zhang authored Jun 05, 2025
```
Signed-off-by: Jerry Zhang <jerryzh168@gmail.com>
```
  c8134bea
05 Jun, 2025 15 commits
- [Model] NemotronH support (#18863) · cb6d572e
  Luis Vega authored Jun 05, 2025
```
Signed-off-by: Luis Vega <2478335+vegaluisjose@users.noreply.github.com>
Co-authored-by: Luis Vega <2478335+vegaluisjose@users.noreply.github.com>
```
  cb6d572e
- [V1] Use FlashInfer by default on Blackwell GPUs (#19118) · 87360308
  Michael Goin authored Jun 05, 2025
  
  87360308
- [P/D][NixlConnector] Enable FlashInfer backend (#19090) · 9ef9173c
  Nicolò Lucchesi authored Jun 05, 2025
  
  9ef9173c
- [MISC][Bugfix] Use less CPU when message queue has been empty for some time (#16226) · 85e2b7bb
  Povilas Kanapickas authored Jun 05, 2025
```
Signed-off-by: Povilas Kanapickas <povilas@radix.lt>
```
  85e2b7bb
- [Hardware][NVIDIA] FP4 MoE kernel optimization (#19110) · 61059bee
  Chiyue Wei authored Jun 05, 2025
```
Signed-off-by: Chiyue Wei <chiyuew@nvidia.com>
Co-authored-by: Chiyue Wei <chiyuew@nvidia.com>
```
  61059bee
- Add H20-3e fused MoE kernel tuning configs for DeepSeek-R1/V3 (#19205) · ec89524f
  Xu Wenqing authored Jun 06, 2025
  
  ec89524f
- [mistral_common] Add v11 tokenizer (#19193) · f20f9f06
  Patrick von Platen authored Jun 05, 2025
```
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
```
  f20f9f06
- [Bugfix] properly catch PIL-related errors for vision models when incorrect... · 9bc8bb07
  Guillaume Calmettes authored Jun 05, 2025
```
[Bugfix] properly catch PIL-related errors for vision models when incorrect data urls are provided (#19202)
Signed-off-by: Guillaume Calmettes <gcalmettes@scaleway.com>
```
  9bc8bb07
- [Frontend] improve vllm run-batch --help display (#19187) · 1aeb925f
  Reid authored Jun 05, 2025
```
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
```
  1aeb925f
- [Misc] Do not override NCCL_CUMEM_ENABLE if set explicitly (#19105) · 188a4590
  22quinn authored Jun 05, 2025
```
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
```
  188a4590
- [Misc] Remove unnecessary fallback to prefill-decode attention (#19138) · 18093084
  vllmellm authored Jun 05, 2025
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
  18093084
- [Bugfix]: Fix the incompatibility issue with stream when Thinking is disabled (#19135) · 8fc57501
  Chauncey authored Jun 05, 2025
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
  8fc57501
- [BugFix][Minor] Fix full cuda graph bug when max_num_seqs < 512 (#19171) · af7fc84f
  Woosuk Kwon authored Jun 04, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  af7fc84f
- [Bugfix] Fix port handling in make_zmq_path (#19117) · a408820f
  Michael Goin authored Jun 04, 2025
  
  a408820f
- [Bugfix][Nixl] Fix full prefix cache hit bug (#18632) · c56ed8bb
  Robert Shaw authored Jun 04, 2025
```
Signed-off-by: rshaw@neuralmagic.com <robertgshaw2@gmail.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  c56ed8bb
04 Jun, 2025 7 commits
- [P/D] Heterogeneous TP (#18833) · b2fac671
  Nicolò Lucchesi authored Jun 05, 2025
```
Signed-off-by: nicklucche <nlucches@redhat.com>
```
  b2fac671
- [Misc] refactor: simplify EngineCoreClient.make_async_mp_client in AsyncLLM (#18817) · 23027e2d
  CYJiang authored Jun 05, 2025
```
Signed-off-by: googs1025 <googs1025@gmail.com>
```
  23027e2d
- [Kernel] Integrate batched/masked deepgemm kernel (#19111) · c3fd4d66
  Varun Sundar Rabindranath authored Jun 04, 2025
```
Signed-off-by: Varun <vsundarr@redhat.com>
Co-authored-by: Varun <vsundarr@redhat.com>
```
  c3fd4d66
- [Bugfix] fix v1 cpu worker fails on macOS (#19121) · ef3f98b5
  Kebe authored Jun 05, 2025
  
  ef3f98b5
- [Bugfix][EP+DP] Fix internode check (#19112) · d459fae0
  Tyler Michael Smith authored Jun 04, 2025
```
Signed-off-by: Tyler Michael Smith <tysmith@redhat.com>
```
  d459fae0
- Allow AsyncLLMEngine.generate to target a specific DP rank (#19102) · c8dcc159
  jmswen authored Jun 04, 2025
```
Signed-off-by: Jon Swenson <jmswen@gmail.com>
```
  c8dcc159
- Sm100 blockwise fp8 swap ab (#18564) · 5f2cd251
  Lain authored Jun 04, 2025
  
  5f2cd251