OneHub AI Gateway

OpenAI-compatible endpoints for language, embedding, image, and audio models. Authenticate with your OneHub API key.

Docs

Getting started

Get a key and hit the chat completions API in minutes.

How keys are scoped, hashed, and presented in headers.

Point any OpenAI-compatible client at the OneHub base URL.

Provider catalog and which plans unlock each model.

Per-plan limits and the rate-limit response headers.

Multi-turn conversations with any supported model.

Token-by-token chunks in OpenAI chat format.

The full catalog filtered to your plan.

Vectorize text for search and RAG workloads.

FLUX and GPT Image 2 from a single endpoint.

Raw audio bytes from the OpenAI TTS models.

Whisper transcription via multipart or JSON.

USD wallet, automatic deduction, and top-ups.

Signed delivery of usage, billing, and key events.

OpenAI-style errors and common HTTP status codes.