A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 04:30 UTC
main
[Misc] Move some multimodal utils to modality-specific modules (#11494)
Cyrus Leung committed almost 2 years ago
[Doc] Improve GitHub links (#11491)
Cyrus Leung committed almost 2 years ago
[Frontend] Enable decord to load video from base64 (#11492)
Cyrus Leung committed almost 2 years ago
[Misc] Update disaggregation benchmark scripts and test logs (#11456)
Jiaxin Shan committed almost 2 years ago
[V1] Unify VLLM_ENABLE_V1_MULTIPROCESSING handling in RayExecutor (#11472)
Rui Qiao committed almost 2 years ago
[Model] Automatic conversion of classification and reward models (#11469)
Cyrus Leung committed almost 2 years ago
[Bugfix] Fix issues in CPU build Dockerfile. Fixes #9182 (#11435)
Yuan Tang committed almost 2 years ago
[Misc] Move weights mapper (#11443)
Jee Jee Li committed almost 2 years ago
[attn][tiny fix] fix attn backend in MultiHeadAttention (#11463)
Mengqing Cao committed almost 2 years ago
[OpenVINO] Fixed installation conflicts (#11458)
Ilya Lavrenov committed almost 2 years ago
[Bugfix][Hardware][CPU] Fix CPU `input_positions` creation for text-only inputs with mrope (#11434)
Isotr0py committed almost 2 years ago
[Bugfix] Fix Qwen2-VL LoRA weight loading (#11430)
Jee Jee Li committed almost 2 years ago
[Frontend] Online Pooling API (#11457)
Cyrus Leung committed almost 2 years ago
[Misc]Suppress irrelevant exception stack trace information when CUDA… (#11438)
dpxa committed almost 2 years ago
Rui Qiao committed almost 2 years ago
[Docs] Convert rST to MyST (Markdown) (#11145)
Rafael Vasquez committed almost 2 years ago
[Doc] Fix typo in the help message of '--guided-decoding-backend' (#11440)
yansh97 committed almost 2 years ago
[Bugfix] Add kv cache scales to gemma2.py (#11269)
Michael Goin committed almost 2 years ago
[CI] Expand OpenAI test_chat.py guided decoding tests (#11048)
Michael Goin committed almost 2 years ago
[Misc] add w8a8 asym models (#11075)
Dipika Sikka committed almost 2 years ago