vLLM
Inference engine with an HTTP server for supported model capabilities.
Official site ↗At a glance
- Organization
- vLLM
- Type
- model inference engine
- Access
- open source
- Deployment
- self hosted
- Lifecycle
- active
When to consider it
- When a supported model needs online inference serving
Considerations
- Endpoint behavior and resource needs vary by model and deployment.
Connections
- In areaComposition & Runtime
- Uses approachModel serving
- Supports functionServe model inference
Sources
Last reviewed 2026-09-30