Runtimes · Inference server

vLLM

High-throughput, memory-efficient serving engine for LLMs with an OpenAI-compatible server.

Facts

Website: vllm.ai
Repository: vllm-project/vllm
Open source: Yes, Apache-2.0 Verified · Source
Pricing: Free: Free Verified · Source
Platforms: macOS Sourced · Source
Added · checked: 2026-10-13 · 2026-10-13
Sponsored: No. Nobody paid for this listing or its position.

Works with local models

Ollama: Not yet confirmed Not yet measured
llama.cpp: Yes Sourced · Source
LM Studio: Not yet confirmed Not yet measured
OpenAI-compatible: Yes Sourced · Source
MCP: Not yet confirmed Not yet measured

Summary: Our one-line paraphrase of the official description. Source.
Logo: trademark of its owner, used only to identify vLLM. Source (favicon svg): https://vllm.ai/vLLM-Logo.svg, checked 2026-10-13.
Machine-readable: Markdown · JSON