Introduction
Developer documentation for the Venna gateway, Console, Venna Code, and node network.
Venna is an OpenAI-compatible inference gateway served by a decentralized node network.
Point any OpenAI client at https://gateway.venna.net/v1, authenticate, and run the chat,
embeddings, audio, image, and video calls you already write.
Text-to-speech and transcription, image generation and editing, and async video generation are live alongside chat and embeddings — see the audio, images, and video guides for request shapes and metering. Models can also call your own functions through tooling.
Start here
Quickstart
Make your first authenticated gateway call in minutes.
Authentication
Workspace keys, plan JWTs, and x402 — when to use each.
API Reference
OpenAI-compatible endpoints, generated from the live spec.
Products
The gateway, Console, Venna Code, and the node network.
Venna Code
The desktop agentic coding mode and its extension points.
Run a Node
Serve inference on your own hardware and earn per token.