Commits · 1cf0719ebdcd67ecdd5801da9154e56683fa7360 · OpenDAS / vllm_cscc

26 Apr, 2025 1 commit
- [Minor][Spec Decode] Add use_eagle to SpeculativeConfig (#17213) · 1cf0719e
  Woosuk Kwon authored Apr 25, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  1cf0719e
25 Apr, 2025 4 commits
- [V1][Spec Decode] EAGLE-3 Support (#16937) · a0e619e6
  Benjamin Chislett authored Apr 25, 2025
```
Signed-off-by: Bryan Lu <yuzhelu@amazon.com>
Signed-off-by: Benjamin Chislett <benjamin.chislett@centml.ai>
Co-authored-by: Bryan Lu <yuzhelu@amazon.com>
```
  a0e619e6
- [V1] Move usage stats to worker and start logging TPU hardware (#16211) · 48cb2109
  Daniel Li authored Apr 25, 2025
  
  48cb2109
- Only turn on FastIncrementalDetokenizer when tokenizers >= 0.21.1 (#17158) · fc966e9c
  Lu Fang authored Apr 25, 2025
  
  fc966e9c
- [Bugfix] remove fallback in guided_json (int range, patterns) (#16725) · 6aae216b
  Sangyeon Cho authored Apr 25, 2025
```
Signed-off-by: csy1204 <josang1204@gmail.com>
Co-authored-by: 조상연[플레이스 AI] <sang-yeon.cho@navercorp.com>
```
  6aae216b
24 Apr, 2025 8 commits
- Add collective_rpc to llm engine (#16999) · fe921763
  Yinghai Lu authored Apr 24, 2025
```
Signed-off-by: Yinghai Lu <yinghai@thinkingmachines.ai>
```
  fe921763
- [V1][Spec Decoding] Add num_drafts and num_accepted_tokens_per_position metrics (#16665) · 340d7b1b
  Mark McLoughlin authored Apr 24, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  340d7b1b
- [V1][Structured Output] Clear xgrammar compiler object when engine core shut... · b724afe3
  Shanshan Shen authored Apr 24, 2025
```
[V1][Structured Output] Clear xgrammar compiler object when engine core shut down to avoid nanobind leaked warning (#16954)
Signed-off-by: shen-shanshan <467638484@qq.com>
```
  b724afe3
- Improve static type checking in `LoRAModelRunnerMixin` (#17104) · 21f4f1c9
  Harry Mellor authored Apr 24, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  21f4f1c9
- [V1][PP] Optimization: continue scheduling prefill chunks (#17080) · c0dfd975
  Rui Qiao authored Apr 24, 2025
```
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
```
  c0dfd975
- Simplify `TokenizerGroup` (#16790) · 0a05ed57
  Harry Mellor authored Apr 24, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  0a05ed57
- [Chore] Remove Sampler from Model Code (#17084) · b411418f
  Woosuk Kwon authored Apr 24, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  b411418f
- [Bugfix] Enable V1 usage stats (#16986) · ed50f466
  Michael Goin authored Apr 23, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  ed50f466
23 Apr, 2025 9 commits
- [V1][Spec Decode] Always use argmax for sampling draft tokens (#16899) · 41fb013d
  Woosuk Kwon authored Apr 23, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  41fb013d
- [BugFix][V1] Fix int32 token index overflow when preparing input ids (#16806) · 32d4b669
  Yong Hoon Shin authored Apr 23, 2025
  
  32d4b669
- [Frontend] Support guidance:no-additional-properties for compatibility with xgrammar (#15949) · 3cde34a4
  Travis Johnson authored Apr 23, 2025
```
Signed-off-by: Travis Johnson <tsjohnso@us.ibm.com>
```
  3cde34a4
- Use `@property` and private field for `data_parallel_rank_local` (#17053) · bdb36603
  Harry Mellor authored Apr 23, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  bdb36603
- Ensure that `pid` passed to `kill_process_tree` is `int` for `mypy` (#17051) · 53c0fa1e
  Harry Mellor authored Apr 23, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  53c0fa1e
- [BugFix] llama4 fa3 fix - RuntimeError: scheduler_metadata must have shape (metadata_size) (#16998) · d0da99fb
  Lucas Wilkinson authored Apr 23, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
```
  d0da99fb
- [V1] Avoid socket errors during shutdown when requests are in in-flight (#16807) · b2f195c4
  Nick Hill authored Apr 22, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  b2f195c4
- [V1][DP] More robust DP/EP dummy request coordination (#16277) · 1e013fa3
  Nick Hill authored Apr 22, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  1e013fa3
- [Core][V1][TPU] Enable structured decoding on TPU V1 (#16499) · 83d93371
  Chenyaaang authored Apr 22, 2025
```
Signed-off-by: Chenyaaang <chenyangli@google.com>
```
  83d93371
22 Apr, 2025 10 commits
- [BugFix] Remove default multiproc executor `collective_rpc` timeout (#17000) · 5175b884
  Nick Hill authored Apr 22, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  5175b884
- [BugFix] Fix incremental detokenization perf issue (#16963) · e4d61442
  Nick Hill authored Apr 22, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  e4d61442
- [V1] Remove pre-allocation for KV cache (#16941) · c4ab9f3e
  Woosuk Kwon authored Apr 22, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  c4ab9f3e
- [Bugfix] Fix the issue where llm.generate cannot be called repeatedly after... · acba33a0
  Chauncey authored Apr 22, 2025
```
[Bugfix] Fix the issue where llm.generate cannot be called repeatedly after setting GuidedDecodingParams (#16767)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
```
  acba33a0
- [Perf] Optimize `_update_states` for GPU model runner (#16910) · a114bf20
  SnowCharm authored Apr 22, 2025
```
Signed-off-by: snowcharm <snowcharmqq@gmail.com>
```
  a114bf20
- [Bugfix]: fix issue with n>1 sampling on v1 requests overriding each other (#16863) · 0e425449
  Jeffrey Li authored Apr 21, 2025
```
Signed-off-by: Jeffrey Li <jeffrey.dot.li@gmail.com>
```
  0e425449
- [BugFix][Spec Decode] No in-place update to draft probs (#16952) · 1311913f
  Woosuk Kwon authored Apr 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  1311913f
- [TPU][V1] Enable Top-P (#16843) · fa3bba2a
  Nicolò Lucchesi authored Apr 22, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
```
  fa3bba2a
- [V1] V1 FlashInfer Attention (#16684) · 986537f1
  Michael Goin authored Apr 21, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Aurick Qiao <qiao@aurick.net>
```
  986537f1
- [TPU][V1] Capture multimodal encoder during model compilation (#15051) · 21020752
  Nicolò Lucchesi authored Apr 22, 2025
```
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Siyuan Liu <lsiyuan@google.com>
```
  21020752
21 Apr, 2025 3 commits
- [TPU][V1] Implicitly adjust page size when there's SMEM OOM (#16871) · 471fe656
  Chengji Yao authored Apr 21, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@google.com>
```
  471fe656
- [V1][Spec Decode] Handle draft tokens beyond max_model_len (#16087) · 3a0fba5c
  Woosuk Kwon authored Apr 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  3a0fba5c
- Restore buffers when wake up from level 2 sleep (#16564) (#16889) · d41faaf9
  Han Zhang authored Apr 21, 2025
```
Signed-off-by: Han <zh950713@gmail.com>
```
  d41faaf9
19 Apr, 2025 3 commits

Serialize tensors using int8 views (#16866) · 87aaadef

Staszek Paśko authored Apr 19, 2025


Signed-off-by: Staszek Pasko <staszek@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>

87aaadef

[V1][Misc] stop update prefix cache stats when logs_stats is disabled (#16460) · d9737ca1
vie-serendipity authored Apr 19, 2025
```
Signed-off-by: vie-serendipity <2733147505@qq.com>
```
d9737ca1

[Model][VLM] Add Qwen2.5-Omni model support (thinker only) (#15130) · 2c1bd848

Yang Fan authored Apr 19, 2025


Signed-off-by: fyabc <suyang.fy@alibaba-inc.com>
Signed-off-by: Roger Wang <ywang@roblox.com>
Co-authored-by: Roger Wang <136131678+ywang96@users.noreply.github.com>
Co-authored-by: Roger Wang <ywang@roblox.com>
Co-authored-by: Xiong Wang <wangxiongts@163.com>

2c1bd848

18 Apr, 2025 1 commit
- [V1][Structured Output] Minor modification to `_validate_structured_output()` (#16748) · 30ed81b7
  Shanshan Shen authored Apr 18, 2025
```
Signed-off-by: shen-shanshan <467638484@qq.com>
```
  30ed81b7
17 Apr, 2025 1 commit
- [Attention] Update to lastest FA3 code (#13111) · 183dad7a
  Lucas Wilkinson authored Apr 17, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
```
  183dad7a