vllm-project/

vllm

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

Pythonamdblackwellcudadeepseekdeepseek-v3+15 more

Stars

71.9k

+53 today+208 /wk+787 /mo

Forks

13.9k

Issues

3.6k

Watchers

71.9k

Star History

LicenseApache-2.0

CreatedFeb 9, 2023

Last push2h ago

Homepagevllm.ai

Date	Stars	Forks	Issues
Mar 2, 2026	71.7k	13.9k	3.6k

vllm-project/vllm — Star Growth & Stats | GitHubVC by AttentionVC