OmniRelay supports live text inference in enabled workspaces. Query
GET /v1/models with your own key for callable IDs, capabilities and customer prices. Registration and public catalog visibility do not enable live access. See Service boundaries.OpenAI-compatible API
Keep your existing OpenAI SDK. Change the base URL and the API key, nothing else.
One gateway, many models
Enabled GPT, Claude, DeepSeek and GLM text models through one OpenAI-compatible endpoint.
Usage and billing built in
Request metadata and settled costs are recorded; wallet movements distinguish cash and trial credits.
Quick Start
First API call
Create a key and send a chat completion in under five minutes.
Authentication
How API keys are issued, stored and revoked.
Overview
Unified gateway
What happens between your request and the model.
Rate limits
Request and payload limits, and how to handle
429.List models
Discover the models enabled in your workspace.
Chat completions
Parameters, examples and unsupported features.
API Reference
Explore the API
Every endpoint with schemas, examples and an interactive playground.
User manual
Accounts, keys, models, playground, usage, billing and troubleshooting.