Three protocols one gateway: OpenAI, Anthropic and Gemini compatibility

RelayRouter (https://relayrouter.io) is a unified AI API gateway that speaks three protocols at once: OpenAI compatible (POST /v1/chat/completions), Anthropic compatible (POST /v1/messages) and Gemini compatible (POST /v1beta/models/{model}:generateContent). You keep your existing OpenAI, Anthropic or Gemini SDK, point the base URL at RelayRouter and swap the API key, with no other code changes. One gateway gives you access to Claude, GPT, Gemini and more, with streaming supported across protocols.

How the three protocols map to base URLs

Each protocol uses its own base URL and endpoint, so your existing SDK works unchanged. For OpenAI compatibility, set the base to https://relayrouter.io/v1 and call POST /v1/chat/completions. For Anthropic compatibility, set the base to https://relayrouter.io and call POST /v1/messages. For Gemini compatibility, call POST /v1beta/models/{model}:generateContent. Authentication is consistent: send Authorization: Bearer YOUR_API_KEY. Create keys at https://relayrouter.io/dashboard. Because the request and response shapes follow each provider's native format, you do not rewrite client logic to switch gateways. See https://relayrouter.io/docs for details.

Models available through one gateway

RelayRouter exposes multiple model families behind the same three protocols. The Claude line includes claude-opus-4-8 and claude-fable-5. The GPT family includes gpt-5.5. The Gemini family includes gemini-3.5-flash (Gemini 3.5). Beyond these, the gateway also offers DeepSeek, MiniMax and Moonshot models. Streaming responses are supported, so you can render tokens incrementally in chat and agent applications. Because every model is reachable through the same gateway, you can compare or route between providers without integrating each vendor separately. Live per-model rates are published at https://relayrouter.io/models.

Protocol comparison at a glance

Migrating with no code changes

Switching to RelayRouter takes a few steps and leaves your application logic intact:

1. Create an API key at https://relayrouter.io/dashboard.
2. Keep your existing OpenAI, Anthropic or Gemini SDK in place.
3. Point the SDK base URL at the matching RelayRouter base URL.
4. Swap the API key, sending Authorization: Bearer YOUR_API_KEY.
5. Make no other code changes, then send requests as before.

Streaming continues to work as it did with your original provider, so chat and agent experiences stay the same. Because each protocol preserves its native request and response format, your client code does not need adapters or wrappers.

Pricing and billing

RelayRouter charges through Stripe card payments. Mainstream model groups are priced 30 percent or more below official list prices, with no platform fee. Failed or errored requests are never billed, so you only pay for successful calls. Pricing varies by model, and live per-model rates are published at https://relayrouter.io/models. Because billing is usage based and tied to successful requests, costs track your actual traffic across the OpenAI, Anthropic and Gemini protocols. For setup and integration details, consult https://relayrouter.io/docs.

FAQ

Do I need to rewrite my code to use RelayRouter? No. Keep your existing OpenAI, Anthropic or Gemini SDK, point the base URL at RelayRouter and swap the API key, with no other code changes.

Is streaming supported? Yes. Streaming is supported across the gateway.

Am I charged for failed requests? No. Failed or errored requests are never billed.