vllm-project/vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

Stars Over Time

87,303 stars by
VTn
PythonApache License 2.0amdblackwellcudadeepseekdeepseek-v3gptgpt-ossinferencekimillamallmllm-servingmodel-servingmoeopenaipytorchqwenqwen3tputransformer

Growth

HOT
Last 30 days+981 stars

Owner

Notable Stargazers

Building vllm-project/vllm?

We track who's paying attention — companies, roles, and the developers most likely to contribute or adopt.

Get in touch