Commits · 3de2b1eafb12e420c563cb7153d4d2f0e8451ca9 · OpenDAS / vllm_cscc · GitLab

10 Jan, 2025 1 commit
- [VLM] Enable tokenized inputs for merged multi-modal processor (#11900) · b844b99a
  Cyrus Leung authored Jan 10, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  b844b99a
09 Jan, 2025 7 commits
- [platform] Allow platform specify attention backend (#11609) · 405eb8e3
  wangxiyuan authored Jan 09, 2025
```
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: Mengqing Cao <cmq0113@163.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>
```
  405eb8e3
- [Doc] Add model development API Reference (#11884) · 65097ca0
  Cyrus Leung authored Jan 09, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  65097ca0
- [Bugfix] fix beam search input errors and latency benchmark script (#11875) · 1d967acb
  Ye (Charlotte) Qi authored Jan 09, 2025
```
Signed-off-by: Ye Qi <yeq@meta.com>
Co-authored-by: yeq <yeq@devgpu004.lla3.facebook.com>
```
  1d967acb
- [Bugfix] Override dunder methods of placeholder modules (#11882) · 0bd1ff43
  Cyrus Leung authored Jan 09, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  0bd1ff43
- [perf]fix current stream (#11870) · 310aca88
  youkaichao authored Jan 09, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  310aca88
- [Misc] Move `print_*_once` from utils to logger (#11298) · d848800e
  Cyrus Leung authored Jan 09, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Maxime Fournioux <55544262+mfournioux@users.noreply.github.com>
Co-authored-by: Maxime Fournioux <55544262+mfournioux@users.noreply.github.com>
```
  d848800e
- treat do_lower_case in the same way as the sentence-transformers library (#11815) · 1fe554ba
  Maximilien de Bayser authored Jan 09, 2025
```
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
```
  1fe554ba
08 Jan, 2025 18 commits
- [Kernel][Triton][AMD] Use block size heuristic for avg 2.8x speedup for int8 models (#11698) · 526de822
  rasmith authored Jan 08, 2025
```
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
```
  526de822
- [TPU][Quantization] TPU `W8A8` (#11785) · 56fe4c29
  Robert Shaw authored Jan 08, 2025
```
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  56fe4c29
- [Misc]add some explanations for BlockHashType (#11847) · 47de8821
  WangErXiao authored Jan 09, 2025
  
  47de8821
- [Doc] Expand Multimodal API Reference (#11852) · 5984499e
  Cyrus Leung authored Jan 09, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  5984499e
- [Misc] Move some model utils into vision file (#11848) · ca47e176
  Cyrus Leung authored Jan 09, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  ca47e176
- [Bugfix][XPU] fix silu_and_mul (#11823) · 78f4590b
  Yan Ma authored Jan 09, 2025
```
Signed-off-by: yan ma <yan.ma@intel.com>
```
  78f4590b
- [CI/Build][Bugfix] Fix CPU CI image clean up (#11836) · 2f702498
  Li, Jiang authored Jan 08, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  2f702498
- [Doc][4/N] Reorganize API Reference (#11843) · 6cd40a5b
  Cyrus Leung authored Jan 08, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  6cd40a5b
- [Doc] Move examples into categories (#11840) · aba8d6ee
  Harry Mellor authored Jan 08, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  aba8d6ee
- [VLM] Reorganize profiling/processing-related code (#11812) · 2a0596bc
  Cyrus Leung authored Jan 08, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  2a0596bc
- [torch.compile] consider relevant code in compilation cache (#11614) · f1214117
  youkaichao authored Jan 08, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  f1214117
- [Hardware][Apple] Native support for macOS Apple Silicon (#11696) · cfd3219f
  Wallas Henrique authored Jan 08, 2025
```
Signed-off-by: Wallas Santos <wallashss@ibm.com>
Co-authored-by: Michael Goin <michael@neuralmagic.com>
```
  cfd3219f
- [misc] improve memory profiling (#11809) · 889e662e
  youkaichao authored Jan 08, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  889e662e
- [Bug] Fix pickling of `ModelConfig` when RunAI Model Streamer is used (#11825) · ef68eb28
  Cyrus Leung authored Jan 08, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  ef68eb28
- [Bugfix] Add checks for LoRA and CPU offload (#11810) · f645eb69
  Jee Jee Li authored Jan 08, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  f645eb69
- Remove the duplicate imports of MultiModalKwargs and PlaceholderRange… (#11824) · dc71af0a
  WangErXiao authored Jan 08, 2025
  
  dc71af0a
- [Bugfix] Fix image input for Pixtral-HF (#11741) · 91445c7b
  Cyrus Leung authored Jan 08, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  91445c7b
- [Bugfix] Significant performance drop on CPUs with --num-scheduler-steps > 1 (#11794) · a4e2b268
  Jie Fu (傅杰) authored Jan 08, 2025
  
  a4e2b268
07 Jan, 2025 10 commits
- [Bugfix] update the prefix for qwen2 (#11795) · c994223d
  jiangjiadi authored Jan 08, 2025
```
Co-authored-by: jiadi.jjd <jiadi.jjd@antgroup.com>
```
  c994223d
- [optimization] remove python function call for custom op (#11750) · 869579a7
  youkaichao authored Jan 08, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  869579a7
- [V1] Support audio language models on V1 (#11733) · 2de197bd
  Roger Wang authored Jan 07, 2025
```
Signed-off-by: Roger Wang <ywang@roblox.com>
```
  2de197bd
- [Bugfix] Comprehensively test and fix LLaVA-NeXT feature size calculation (#11800) · 8f37be38
  Cyrus Leung authored Jan 07, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  8f37be38
- fix init error for MessageQueue when n_local_reader is zero (#11768) · e512f76a
  XiaobingZhang authored Jan 07, 2025
  
  e512f76a
- [Bugfix] Update attention interface in `Whisper` (#11784) · 0f3f3c86
  Roger Wang authored Jan 06, 2025
```
Signed-off-by: Roger Wang <ywang@roblox.com>
```
  0f3f3c86
- [Kernel][LoRA]Punica prefill kernels fusion (#11234) · b2785579
  Jee Jee Li authored Jan 07, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Signed-off-by: Abatom <abzhonghua@gmail.com>
Co-authored-by: Zhonghua Deng <abatom@163.com>
```
  b2785579
- [XPU] Make pp group initilized for pipeline-parallelism (#11648) · d93d2d74
  YiSheng5 authored Jan 07, 2025
```
Signed-off-by: yisheng <yi.sheng@intel.com>
```
  d93d2d74
- [Model] Future-proof Qwen2-Audio multi-modal processor (#11776) · d0169e1b
  Cyrus Leung authored Jan 07, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d0169e1b
- [Bugfix] Fix LLaVA-NeXT feature size precision error (for real) (#11772) · 08fb75c7
  Cyrus Leung authored Jan 07, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  08fb75c7
06 Jan, 2025 4 commits
- [V1] Extend beyond image modality and support mixed-modality inference with... · 91b361ae
  Roger Wang authored Jan 06, 2025
```
[V1] Extend beyond image modality and support mixed-modality inference with Llava-OneVision (#11685)
Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  91b361ae
- [Kernel] Move attn_type to Attention.__init__() (#11690) · e20c92bb
  Chen Zhang authored Jan 07, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  e20c92bb
- [Bugfix][V1] Fix molmo text-only inputs (#11676) · 32c9eff2
  Jee Jee Li authored Jan 06, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  32c9eff2
- [Bugfix] Fix max image size for LLaVA-Onevision (#11769) · 9279b9f8
  Roger Wang authored Jan 06, 2025
```
Signed-off-by: Roger Wang <ywang@roblox.com>
```
  9279b9f8