# vLLM

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

Canonical: https://thelocalbench.com/tools/vllm · Data as of 2026-10-12 · The Local Bench (thelocalbench.com), data as of 2026-10-12, CC BY 4.0

Flags: [verified] checked against primary source · [sourced] cited · [estimate] formula · [sample] placeholder

- Category: Runtimes / Inference server
- Website: https://vllm.ai
- Repository: https://github.com/vllm-project/vllm
- Open source: yes, Apache-2.0 [verified] https://github.com/vllm-project/vllm/blob/HEAD/LICENSE
- Pricing: free, starting $0 [verified] https://github.com/vllm-project/vllm
- Platforms: macOS [sourced] https://github.com/vllm-project/vllm#readme
- Added 2026-10-13, last verified 2026-10-13
- Sponsored: no

## Works with local models

| Integration | Status | Flag | Source |
|---|---|---|---|
| Ollama | not yet confirmed | [not_measured] |  |
| llama.cpp | yes | [sourced] | https://github.com/vllm-project/vllm#readme |
| LM Studio | not yet confirmed | [not_measured] |  |
| OpenAI-compatible | yes | [sourced] | https://github.com/vllm-project/vllm#readme |
| MCP | not yet confirmed | [not_measured] |  |

JSON: https://thelocalbench.com/api/v1/tools/vllm.json

Logo: trademark of its owner, shown to identify the tool (source: favicon-svg, https://vllm.ai/vLLM-Logo.svg).