Introduction
What Model Gate is, how billing works, and which API formats are available.
Introduction
What is Model Gate?
Model Gate is an AI gateway with one API endpoint supporting both OpenAI- and Anthropic-compatible request formats. A single API key provides access to every enabled model.
How it works
1. Create an API key. 2. Add funds to your account balance. 3. Send requests to the Model Gate endpoint. 4. The gateway routes each request to the selected provider. 5. The final request cost is deducted from your USD balance. 6. Review requests, token usage, cost, and transactions in the control panel.
Billing model
Billing uses a funded USD balance. Each request is settled after completion from actual model usage and the configured per-million-token prices. API keys can have independent spend, RPM, concurrency, and model restrictions.
Supported formats
| Format | Endpoint | Compatible clients |
|---|---|---|
| OpenAI Chat Completions | POST /v1/chat/completions | OpenAI SDK, LangChain, and OpenAI-compatible clients |
| Anthropic Messages | POST /v1/messages | Anthropic SDK and Claude API clients |
Base URL
https://api.model-gate.comOpenAI-compatible clients normally use https://api.model-gate.com/v1 as their Base URL.
Ready-made configurations are available in the Tools section. Start with Claude Code or Codex CLI.