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, 10:45 UTC
main
[Model] Make llama3.2 support multiple and interleaved images (#9095)
Xiang Xu committed about 2 years ago
[Frontend] merge beam search implementations (#9296)
Brendan Wong committed about 2 years ago
[TPU] Fix TPU SMEM OOM by Pallas paged attention kernel (#9350)
Woosuk Kwon committed about 2 years ago
[Docs] Remove PDF build from Readtehdocs (#9347)
Simon Mo committed about 2 years ago
[CI/Build] setuptools-scm fixes (#8900)
Daniele committed about 2 years ago
[Hardware][intel GPU] add async output process for xpu (#8897)
Kunshang Ji committed about 2 years ago
[Model] Molmo vLLM Integration (#9016)
Reza Salehi committed about 2 years ago
[Bugfix] Bandaid fix for speculative decoding tests (#9327)
Tyler Michael Smith committed about 2 years ago
[CI] Fix merge conflict (#9317)
Lily Liu committed about 2 years ago
[Bugfix]Fix MiniCPM's LoRA bug (#9286)
Jee Jee Li committed about 2 years ago
[Misc][Installation] Improve source installation script and doc (#9309)
Yunmeng committed about 2 years ago
[Bugfix] Fix bug of xformer prefill for encoder-decoder (#9026)
Xiang Xu committed about 2 years ago
[SpecDec] Remove Batch Expansion (2/3) (#9298)
Lily Liu committed about 2 years ago
[BugFix] Fix tool call finish reason in streaming case (#9209)
Maximilien de Bayser committed about 2 years ago
[bugfix] fix f-string for error (#9295)
Prashant Gupta committed about 2 years ago
[Bugfix] Sets `is_first_step_output` for TPUModelRunner (#9202)
Allen Wang committed about 2 years ago
[Bugfix][CI/Build] Fix docker build where CUDA archs < 7.0 are being detected (#9254)
Lucas Wilkinson committed about 2 years ago
[Doc] Compatibility matrix for mutual exclusive features (#8512)
Wallas Henrique committed about 2 years ago
[Doc] Remove outdated comment to avoid misunderstanding (#9287)
homeffjy committed about 2 years ago
[Model] Add GLM-4v support and meet vllm==0.6.2 (#9242)
sixgod committed about 2 years ago