Commits · 09428eeca05aaa59563474d87af7f16f61b20036 · OpenDAS / vllm_cscc · GitLab

14 Jan, 2025 1 commit
- [feat]上传deepseek_v2_lite、qwen2_moe、mixtral 8*7B、8x822B 4个模型的fused_moe_kernel config · 09428eec
  zhuwenwen authored Jan 14, 2025
  
  09428eec
13 Jan, 2025 2 commits
- update alibi · dc5f1906
  zhuwenwen authored Jan 13, 2025
  
  dc5f1906
- fix alibi · 3f946cfb
  zhuwenwen authored Jan 13, 2025
  
  3f946cfb
08 Jan, 2025 1 commit
- resolve transformers version conflicts with lmslim · 0890ebc1
  zhuwenwen authored Jan 08, 2025
  
  0890ebc1
06 Jan, 2025 2 commits
- for qwen2, mixtral and mistral models, remove the warning from _ROCM_SWA_REASON · a9c53382
  zhuwenwen authored Jan 06, 2025
  
  a9c53382
- update qwen2-moe layout and benchmark_throughput.py · 220e6456
  zhuwenwen authored Jan 06, 2025
  
  220e6456
04 Jan, 2025 2 commits
- Merge tag 'v0.6.6.post1' into v0.6.6.post1-dev · 96ae75ad
  zhuwenwen authored Jan 04, 2025
  
  96ae75ad
- update qwen&qwen2-moe and baichuan layout · f9f4a735
  zhuwenwen authored Jan 04, 2025
  
  f9f4a735
03 Jan, 2025 4 commits
- remove gemm and fa pad · ca7f98e5
  zhuwenwen authored Jan 03, 2025
  
  ca7f98e5
- update qwen2 and mixtral layout · 57adffa2
  zhuwenwen authored Jan 03, 2025
  
  57adffa2
- fix run error · 184b50f7
  zhuwenwen authored Jan 03, 2025
  
  184b50f7
- skip ggml_dequantize · 28a8a733
  zhuwenwen authored Jan 03, 2025
  
  28a8a733
02 Jan, 2025 1 commit
- add softcap interface for gemma-2 · f9b567df
  zhuwenwen authored Jan 02, 2025
  
  f9b567df
31 Dec, 2024 1 commit
- remove fatrelu_and_mul · 6dc7aa42
  zhuwenwen authored Dec 31, 2024
  
  6dc7aa42
30 Dec, 2024 2 commits
- fix interface error · 02c3e313
  zhuwenwen authored Dec 30, 2024
  
  02c3e313
- Merge tag 'v0.6.5' into v0.6.5-dev · 4d3a2c28
  zhuwenwen authored Dec 30, 2024
  
  4d3a2c28
27 Dec, 2024 9 commits
- [BugFix] Fix quantization for all other methods (#11547) · 2339d59f
  Robert Shaw authored Dec 27, 2024
  
  2339d59f
- [V1][3/N] API Server: Reduce Task Switching + Handle Abort Properly (#11534) · 1b875a0e
  Robert Shaw authored Dec 27, 2024
  
  1b875a0e
- [misc] fix typing (#11540) · eb881ed0
  youkaichao authored Dec 27, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  eb881ed0
- [CI] Fix broken CI (#11543) · 46d43594
  Robert Shaw authored Dec 26, 2024
  
  46d43594
- [V1] Fix yapf (#11538) · 81b979f2
  Woosuk Kwon authored Dec 27, 2024
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  81b979f2
- [V1] Use FlashInfer Sampling Kernel for Top-P & Top-K Sampling (#11394) · 371d04d3
  Woosuk Kwon authored Dec 27, 2024
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  371d04d3
- Update openai_compatible_server.md (#11536) · 0c0c2015
  Robert Shaw authored Dec 26, 2024
```
Co-authored-by: Simon Mo <simon.mo@hey.com>
```
  0c0c2015
- [Docs] Document Deepseek V3 support (#11535) · 82d24f7a
  Simon Mo authored Dec 26, 2024
```
Signed-off-by: simon-mo <simon.mo@hey.com>
```
  82d24f7a
- Deepseek v3 (#11502) · f49777ba
  Simon Mo authored Dec 26, 2024
```
Signed-off-by: mgoin <michael@neuralmagic.com>
Co-authored-by: mgoin <michael@neuralmagic.com>
Co-authored-by: robertgshaw2-neuralmagic <rshaw@neuralmagic.com>
```
  f49777ba
26 Dec, 2024 11 commits
- [2/N] API Server: Avoid ulimit footgun (#11530) · 55fb97f7
  Robert Shaw authored Dec 26, 2024
  
  55fb97f7
- [Model] [Quantization] Support deepseek_v3 w8a8 fp8 block-wise quantization (#11523) · 2072924d
  Michael Goin authored Dec 26, 2024
```
Signed-off-by: mgoin <michael@neuralmagic.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
Signed-off-by: simon-mo <xmo@berkeley.edu>
Co-authored-by: simon-mo <simon.mo@hey.com>
Co-authored-by: simon-mo <xmo@berkeley.edu>
Co-authored-by: HandH1998 <1335248067@qq.com>
```
  2072924d
- [1/N] API Server (Remove Proxy) (#11529) · 720b10fd
  Robert Shaw authored Dec 26, 2024
  
  720b10fd
- [Doc] Add video example to openai client for multimodal (#11521) · b85a9778
  Isotr0py authored Dec 27, 2024
```
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  b85a9778
- [Misc] Add placeholder module (#11501) · eec906d8
  Cyrus Leung authored Dec 26, 2024
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  eec906d8
- [Model] Modify MolmoForCausalLM MLP (#11510) · f57ee565
  Jee Jee Li authored Dec 26, 2024
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  f57ee565
- [V1] Adding min tokens/repetition/presence/frequence penalties to V1 sampler (#10681) · dcb1a944
  sroy745 authored Dec 26, 2024
```
Signed-off-by: Sourashis Roy <sroy@roblox.com>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  dcb1a944
- [Doc] Add `QVQ` and `QwQ` to the list of supported models (#11509) · 7492a362
  Roger Wang authored Dec 26, 2024
```
Signed-off-by: Roger Wang <ywang@roblox.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  7492a362
- [Misc][LoRA] Fix LoRA weight mapper (#11495) · aa25985b
  Jee Jee Li authored Dec 26, 2024
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  aa25985b
- Mypy checking for vllm/compilation (#11496) · dbeac95d
  Lucas Tucker authored Dec 25, 2024
```
Signed-off-by: lucast2021 <lucast2021@headroyce.org>
Co-authored-by: lucast2021 <lucast2021@headroyce.org>
```
  dbeac95d
- [Misc] Move some multimodal utils to modality-specific modules (#11494) · 51a624bf
  Cyrus Leung authored Dec 26, 2024
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  51a624bf
25 Dec, 2024 4 commits
- [Doc] Improve GitHub links (#11491) · 6ad909fd
  Cyrus Leung authored Dec 26, 2024
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  6ad909fd
- [Frontend] Enable decord to load video from base64 (#11492) · b689ada9
  Cyrus Leung authored Dec 26, 2024
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  b689ada9
- [Misc] Update disaggregation benchmark scripts and test logs (#11456) · fc601665
  Jiaxin Shan authored Dec 24, 2024
```
Signed-off-by: Jiaxin Shan <seedjeffwan@gmail.com>
```
  fc601665
- add torch2.3 · 92ec5d8e
  zhuwenwen authored Dec 25, 2024
  
  92ec5d8e