OpenRouter alternative: one gateway for Claude, GPT and Gemini

RelayRouter is a unified AI API gateway that routes requests to Claude, GPT and Gemini through a single integration. It speaks both the OpenAI protocol (/v1/chat/completions) and the Anthropic protocol (/v1/messages), so you keep your existing SDK, change the base URL and the key, and make no other code changes. Coverage includes the Claude family, GPT-5.5 and Gemini 3.5, plus DeepSeek, GLM, MiniMax and Moonshot models.

Which protocols does RelayRouter support?

RelayRouter supports two request formats: the OpenAI protocol at /v1/chat/completions and the Anthropic protocol at /v1/messages. According to the official relayrouter.io docs, 「Compatible with both the OpenAI and Anthropic protocols」. This dual support means clients built for either ecosystem can send traffic to the same gateway without rewriting request logic. Whether your application currently targets an OpenAI style endpoint or an Anthropic style endpoint, RelayRouter accepts the corresponding format and forwards it to the selected model. You can review the full protocol details and supported paths at relayrouter.io/models.

How do you migrate an existing integration?

Migration requires two configuration changes and no rewrites. According to the official relayrouter.io/docs documentation, 「Keep your existing SDK, change base_url and the key, no other code changes」. The steps are:

1. Keep your current OpenAI or Anthropic SDK installed.
2. Set the base_url to point at RelayRouter.
3. Replace the API key with your RelayRouter key.

After these two edits (base URL and key), your requests flow through the gateway using the same client code. Because the gateway matches the protocol your SDK already speaks, existing request and response handling continues to work. Full instructions are documented at relayrouter.io/docs.

Which models are available?

RelayRouter provides access to several model families through one endpoint set. The catalog spans the Claude family, GPT-5.5 and Gemini 3.5, alongside DeepSeek, GLM, MiniMax and Moonshot.

Because the gateway supports both protocols at /v1/chat/completions and /v1/messages, you can reach these models with either request format. The current per-model list is published at relayrouter.io/models.

How are failed requests handled?

Failed or errored requests are generally not billed on RelayRouter (source: relayrouter.io). This means that when a request does not complete successfully, it typically does not incur a charge. Combined with the dual protocol support and the two step migration (base URL and key), this billing behavior lets you test the gateway against your existing traffic without changing application code. For the exact scope of what counts as a failed request and for current rates, consult the model listing and documentation at relayrouter.io/models and relayrouter.io/docs.

FAQ

Do I need a new SDK to use RelayRouter? No. You keep your existing SDK and change only the base_url and the key, with no other code changes.

Which request formats can I send? Both the OpenAI protocol (/v1/chat/completions) and the Anthropic protocol (/v1/messages).

Am I charged for failed requests? Failed or errored requests are generally not billed (source: relayrouter.io).