vLLM

A high-throughput and memory-efficient inference engine for large language models.

Explore on Agidapp Global