MODEL-RUNNER / VLLM

vLLM

High-throughput GPU serving for open models.

model-runnerfreeself-host