Commits · 47831430cc943cd470d38d27f8c69a5782795ec3 · OpenDAS / vllm_cscc

04 Jan, 2025 10 commits
- [Bugfix][V1] Fix test_kv_cache_utils.py (#11738) · 47831430
  Jee Jee Li authored Jan 05, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  47831430
- [Model] Remove unnecessary weight initialization logic (#11736) · 65c08928
  Cyrus Leung authored Jan 04, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
```
  65c08928
- [Bugfix] Fix precision error in LLaVA-NeXT (#11735) · ba214dff
  Cyrus Leung authored Jan 04, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  ba214dff
- [VLM] Merged multi-modal processors for LLaVA-NeXT-Video and LLaVA-OneVision (#11717) · eed11ebe
  Cyrus Leung authored Jan 04, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  eed11ebe
- [Core][Bugfix] Use correct device to initialize GPU data during CUDA-graph-capture (#11233) · 300acb83
  Yan Burman authored Jan 04, 2025
```
Signed-off-by: Yan Burman <yanburman@users.noreply.github.com>
Signed-off-by: Ido Asraff <idoa@atero.ai>
```
  300acb83
- [V1] Add kv cache utils tests. (#11513) · d91457d5
  xcnick authored Jan 04, 2025
```
Signed-off-by: xcnick <xcnick0412@gmail.com>
```
  d91457d5
- [V1] Add `RayExecutor` support for `AsyncLLM` (api server) (#11712) · fbf25645
  Kunshang Ji authored Jan 04, 2025
  
  fbf25645
- Update bnb.md with example for OpenAI (#11718) · d1d49397
  Alberto Ferrer authored Jan 04, 2025
  
  d1d49397
- Update tool_calling.md (#11701) · 9c93636d
  Hust_YangXian authored Jan 04, 2025
  
  9c93636d
- [V1] log GPU blocks num for MultiprocExecutor (#11656) · e5d7ed0c
  WangErXiao authored Jan 04, 2025
  
  e5d7ed0c
03 Jan, 2025 10 commits
- [V1] Chore: cruft removal (#11724) · ad0d567e
  Robert Shaw authored Jan 03, 2025
  
  ad0d567e
- Update requirements-tpu.txt to support python 3.9 and 3.11 (#11695) · bf0d97d7
  Michael Goin authored Jan 03, 2025
```
Signed-off-by: mgoin <michael@neuralmagic.com>
```
  bf0d97d7
- [Misc]Add BNB quantization for Qwen2VL (#11719) · a655eb30
  Jee Jee Li authored Jan 04, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
```
  a655eb30
- [V1] Improve TP>1 Error Handling + Stack Trace (#11721) · 1543914c
  Robert Shaw authored Jan 03, 2025
```
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  1543914c
- [Bugfix] Fix ColumnParallelLinearWithLoRA slice (#11708) · 61fed92c
  ZincCat authored Jan 03, 2025
```
Signed-off-by: ZincCat <zincchloride@outlook.com>
```
  61fed92c
- [V1] Simplify Shutdown (#11659) · 80c751e7
  Robert Shaw authored Jan 03, 2025
  
  80c751e7
- [Model] Whisper model implementation (#11280) · e1a5c2f0
  Aurick Qiao authored Jan 03, 2025
```
Co-authored-by: Aurick Qiao <aurick.qiao@snowflake.com>
```
  e1a5c2f0
- [perf-benchmark] Fix dependency for steps in benchmark pipeline (#11710) · fd3a62a1
  Kevin H. Luu authored Jan 03, 2025
  
  fd3a62a1
- [Bugfix] Check chain_speculative_sampling before calling it (#11673) · 07064cb1
  Lu Fang authored Jan 02, 2025
```
Signed-off-by: Lu Fang <lufang@fb.com>
```
  07064cb1
- Update default max_num_batch_tokens for chunked prefill (#11694) · 2f1e8e8f
  Sachin Varghese authored Jan 02, 2025
  
  2f1e8e8f
02 Jan, 2025 9 commits
- [Misc] Minimum requirements for SageMaker compatibility (#11576) · 68d37809
  Nathan Azrak authored Jan 03, 2025
  
  68d37809
- Resolve race conditions in Marlin kernel (#11493) · 5dba2575
  wchen61 authored Jan 03, 2025
```
Signed-off-by: wchen61 <wchen61@foxmail.com>
```
  5dba2575
- [Bugfix] Change kv scaling factor by param json on nvidia gpu (#11688) · 187e3299
  bjmsong authored Jan 03, 2025
```
Signed-off-by: bjmsong <bjmsong@126.com>
Co-authored-by: bjmsong <bjmsong@126.com>
```
  187e3299
- [V1][Minor] Optimize token_ids_cpu copy (#11692) · b55ed6ef
  Woosuk Kwon authored Jan 03, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  b55ed6ef
- [Bugfix] Free cross attention block table for preempted-for-recompute sequence group. (#10013) · 2f385183
  Kathy Yu authored Jan 02, 2025
```
Signed-off-by: Kathy Yu <feiyangyu@google.com>
```
  2f385183
- According to vllm.EngineArgs, the name should be distributed_executor_backend (#11689) · 84c35c37
  Chunyang Wen authored Jan 03, 2025
  
  84c35c37
- [VLM] Merged multi-modal processor for LLaVA-NeXT (#11682) · 8c38ee70
  Cyrus Leung authored Jan 03, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  8c38ee70
- [mypy] Pass type checking in vllm/inputs (#11680) · b6087a6b
  Tobias Pitters authored Jan 02, 2025
```
Signed-off-by: Tobias Pitters <tobias.pitters@gmail.com>
```
  b6087a6b
- [VLM][Bugfix] Multi-modal processor compatible with V1 multi-input (#11674) · 23c1b10a
  Cyrus Leung authored Jan 02, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  23c1b10a
01 Jan, 2025 8 commits
- [VLM] Move supported limits and max tokens to merged multi-modal processor (#11669) · a115ac46
  Cyrus Leung authored Jan 01, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
```
  a115ac46
- [V1] Implement Cascade Attention (#11635) · 73001445
  Woosuk Kwon authored Jan 01, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  73001445
- [Doc] Fix typo (#11666) · 6d70198b
  Kazuhiro Serizawa authored Jan 01, 2025
```
Signed-off-by: Kazuhiro Serizawa <nserihiro@gmail.com>
```
  6d70198b
- [Misc] Replace space with - in the file names (#11667) · f962f426
  Lu Fang authored Dec 31, 2024
```
Signed-off-by: Lu Fang <lufang@fb.com>
```
  f962f426
- [Misc] Optimize Qwen2-VL LoRA test (#11663) · 11d8a091
  Jee Jee Li authored Jan 01, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  11d8a091
- [VLM] Add max-count checking in data parser for single image models (#11661) · 365801fe
  Cyrus Leung authored Jan 01, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Roger Wang <ywang@roblox.com>
Co-authored-by: Roger Wang <ywang@roblox.com>
```
  365801fe
- [Bugfix][Refactor] Unify model management in frontend (#11660) · 4db72e57
  Joe Runde authored Dec 31, 2024
```
Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>
```
  4db72e57
- [Benchmark] Add benchmark script for CPU offloading (#11533) · 0c6f9985
  Yihua Cheng authored Dec 31, 2024
```
Signed-off-by: ApostaC <yihua98@uchicago.edu>
Co-authored-by: KuntaiDu <kuntai@uchicago.edu>
```
  0c6f9985
31 Dec, 2024 3 commits

[V1][VLM] V1 support for selected single-image models. (#11632) · e7c7c5e8

Roger Wang authored Dec 31, 2024


Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Isotr0py <2037008807@qq.com>

e7c7c5e8

[V1] Simpify vision block hash for prefix caching by removing offset from hash (#11646) · 8c3230d8
Chen Zhang authored Dec 31, 2024

8c3230d8

[Bugfix] Move the _touch(computed_blocks) call in the allocate_slots method to... · 2c571880

sakunkun authored Dec 31, 2024

[Bugfix] Move the _touch(computed_blocks) call in the allocate_slots method to after the check for allocating new blocks. (#11565)

2c571880