vLLM

Enterprise LLM serving with sub-100ms latency at scale, tensor parallelism, and NVIDIA GPU support...

Open FounderOS desktop