Commits · 228cfbd03fd1ad9b26001817a6d414cc9f2c22ae · OpenDAS / vllm_cscc · GitLab

25 Oct, 2024 5 commits
- [Doc] Improve quickstart documentation (#9256) · 228cfbd0
  Rafael Vasquez authored Oct 25, 2024
```
Signed-off-by: Rafael Vasquez <rafvasq21@gmail.com>
```
  228cfbd0
- [Bugfix] Fix compressed_tensors_moe bad config.strategy (#9677) · ca0d9222
  Michael Goin authored Oct 25, 2024
  
  ca0d9222
- [V1] Support sliding window attention (#9679) · 9645b9f6
  Woosuk Kwon authored Oct 24, 2024
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  9645b9f6
- [Model] add a lora module for granite 3.0 MoE models (#9673) · a6f37218
  Will Johnson authored Oct 25, 2024
  
  a6f37218
- [ci/Build] Skip Chameleon for transformers 4.46.0 on broadcast test #9675 (#9676) · 9f7b4ba8
  Kevin H. Luu authored Oct 24, 2024
  
  9f7b4ba8
24 Oct, 2024 19 commits
- [Bugfix] Remove xformers requirement for Pixtral (#9597) · c91ed47c
  Michael Goin authored Oct 24, 2024
```
Signed-off-by: mgoin <michael@neuralmagic.com>
```
  c91ed47c
- [Performance][Kernel] Fused_moe Performance Improvement (#9384) · 59449095
  Charlie Fu authored Oct 24, 2024
```
Signed-off-by: charlifu <charlifu@amd.com>
```
  59449095
- [Log][Bugfix] Fix default value check for `image_url.detail` (#9663) · e26d37a1
  Michael Goin authored Oct 24, 2024
  
  e26d37a1
- [Model] Compute Llava Next Max Tokens / Dummy Data From Gridpoints (#9650) · 722d46ed
  Alex Brooks authored Oct 24, 2024
```
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
```
  722d46ed
- [CI/Build] Fix VLM test failures when using transformers v4.46 (#9666) · c866e007
  Cyrus Leung authored Oct 25, 2024
  
  c866e007
- [torch.compile] Adding torch compile annotations to some models (#9641) · d27cfbf7
  Yongzao authored Oct 25, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>
```
  d27cfbf7
- Increase operation per run limit for "Close inactive issues and PRs" workflow (#9661) · de662d32
  Harry Mellor authored Oct 24, 2024
```
Signed-off-by: Harry Mellor <hej.mellor@gmail.com>
```
  de662d32
- [Bugfix]Disable the post_norm layer of the vision encoder for LLaVA models (#9653) · f5845496
  litianjian authored Oct 24, 2024
  
  f5845496
- [Doc] Move additional tips/notes to the top (#9647) · b979143d
  Cyrus Leung authored Oct 24, 2024
  
  b979143d
- [torch.compile] expanding support and fix allgather compilation (#9637) · ad6f7805
  Yongzao authored Oct 24, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>
```
  ad6f7805
- [Kernel] add kernel for FATReLU (#9610) · 295a061f
  Jee Jee Li authored Oct 24, 2024
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  295a061f
- [torch.compile] Adding torch compile annotations to some models (#9639) · 8a02cd04
  Yongzao authored Oct 24, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>
```
  8a02cd04
- [core] simplify seq group code (#9569) · 4fdc581f
  youkaichao authored Oct 24, 2024
```
Co-authored-by: Zhuohan Li <zhuohan123@gmail.com>
```
  4fdc581f
- [V1][Bugfix] Clean up requests when aborted (#9629) · 3770071e
  Woosuk Kwon authored Oct 23, 2024
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  3770071e
- [Bugfix] Fix PP for ChatGLM and Molmo (#9422) · 836e8ef6
  Cyrus Leung authored Oct 24, 2024
  
  836e8ef6
- [XPU] avoid triton import for xpu (#9440) · 056a68c7
  Yan Ma authored Oct 24, 2024
```
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  056a68c7
- [Bugfix]: Make chat content text allow type content (#9358) · 33bab410
  Vinay R Damodaran authored Oct 24, 2024
```
Signed-off-by: Vinay Damodaran <vrdn@hey.com>
```
  33bab410
- [Bugfix] Use "vision_model" prefix for MllamaVisionModel (#9628) · b7df53cd
  Michael Goin authored Oct 23, 2024
```
Signed-off-by: mgoin <michael@neuralmagic.com>
```
  b7df53cd
- [Bugfix][Model] Fix Mllama SDPA illegal memory access for batched multi-image (#9626) · bb01f291
  Michael Goin authored Oct 23, 2024
```
Signed-off-by: mgoin <michael@neuralmagic.com>
```
  bb01f291
23 Oct, 2024 16 commits
- [CI/Build] Add bot to close stale issues and PRs (#9436) · b548d7a5
  Russell Bryant authored Oct 23, 2024
  
  b548d7a5
- [Model] Add Qwen2-Audio model support (#9248) · fc6c2746
  Yunfei Chu authored Oct 24, 2024
```
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  fc6c2746
- [Frontend] Enable Online Multi-image Support for MLlama (#9393) · 150b7790
  Alex Brooks authored Oct 23, 2024
```
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  150b7790
- [torch.compile] Adding torch compile annotations to some models (#9614) · 9013e24f
  Yongzao authored Oct 24, 2024
  
  9013e24f
- [Misc] Separate total and output tokens in benchmark_throughput.py (#8914) · fd0e2cfd
  Michael Goin authored Oct 23, 2024
  
  fd0e2cfd
- [Bugfix] Fix divide by zero when serving Mamba models (#9617) · e5ac6a41
  Tyler Michael Smith authored Oct 23, 2024
```
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  e5ac6a41
- [misc] comment to avoid future confusion about baichuan (#9620) · dbdd3b5e
  youkaichao authored Oct 23, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  dbdd3b5e
- [Bugfix] Fix `_init_vision_model` in NVLM_D model (#9611) · e7116c01
  Cyrus Leung authored Oct 23, 2024
```
Co-authored-by: Isotr0py <2037008807@qq.com>
```
  e7116c01
- [Model] Add min_pixels / max_pixels to Qwen2VL as mm_processor_kwargs (#9612) · 31a08f5b
  Alex Brooks authored Oct 23, 2024
```
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
```
  31a08f5b
- [VLM] Enable overriding whether post layernorm is used in vision encoder + fix quant args (#9217) · c18e1a34
  Cyrus Leung authored Oct 23, 2024
```
Co-authored-by: Isotr0py <2037008807@qq.com>
```
  c18e1a34
- [Model] Initialize Florence-2 language backbone support (#9555) · 3ff57ebf
  Isotr0py authored Oct 23, 2024
  
  3ff57ebf
- [Hardware][XPU] using current_platform.is_xpu (#9605) · 2394962d
  Mengqing Cao authored Oct 23, 2024
  
  2394962d
- [Build] Fix `FetchContent` multiple build issue (#9596) · 51c24c97
  Luka Govedič authored Oct 23, 2024
```
Signed-off-by: luka <luka@neuralmagic.com>
```
  51c24c97
- [Model] Support E5-V (#9576) · 831540cf
  Cyrus Leung authored Oct 23, 2024
  
  831540cf
- [Misc] Add an env var VLLM_LOGGING_PREFIX, if set, it will be prepend to all... · 29061ed9
  Flex Wang authored Oct 22, 2024
```
[Misc] Add an env var VLLM_LOGGING_PREFIX, if set, it will be prepend to all logging messages (#9590)
```
  29061ed9
- [Bugfix] Generate exactly input_len tokens in benchmark_throughput (#9592) · 65050a40
  Chen Zhang authored Oct 22, 2024
  
  65050a40