Last released Jun 1, 2024
A high-throughput and memory-efficient inference and serving engine for LLMs
Supported by