# vllm
High-throughput, memory-efficient engine for serving LLMs.
- Rank: #24 of 52 · #5 in Local AI
- Score: 54 (relative; maintenance 64, adoption 70, docs 13; community not measured)
- Safety: Caution — Skill/agent file instructs piping a remote script into a shell: AGENTS.md — curl -LsSf https://astral.sh/uv/install.sh | sh (automated scan of commit 0e5b2ba on 2026-11-02, not a manual audit)
- Stars: 96100
- Last push: 2026-11-02
- License: Apache-2.0
- Repo: https://github.com/vllm-project/vllm
- Fetched: 2026-11-02T06:08:26.085Z