Model not found error on RelayRouter: checking your key's model group

A "model not found" error on RelayRouter usually means the model ID you sent is not in your API key's assigned model group, or the ID string does not match a supported model. RelayRouter covers the Claude family, GPT-5.5, Gemini 3.5, plus DeepSeek, GLM, MiniMax and Moonshot. Confirm the exact model ID against the live list at relayrouter.io/models, verify your key, and resend the request.

Confirm the model ID against the supported list

Start by checking that your requested model string exactly matches a model RelayRouter supports. The supported set spans the Claude family, GPT-5.5, Gemini 3.5, and additional providers including DeepSeek, GLM, MiniMax and Moonshot, published at relayrouter.io/models. A typo, an outdated model name, or a model outside these four provider groups will return "model not found." Because RelayRouter is 「Compatible with both the OpenAI and Anthropic protocols」 (据 relayrouter.io 官方文档), the model ID you pass must correspond to a listed model on whichever protocol endpoint you call. Copy the ID directly from the models page to avoid mismatches.

Check the protocol and base URL your key is using

Verify you are calling the correct protocol endpoint and base URL for your SDK. RelayRouter accepts both the OpenAI path (/v1/chat/completions) and the Anthropic path (/v1/messages), per relayrouter.io/models. If your base URL or endpoint does not match the protocol your SDK expects, a valid model can still be rejected. To migrate correctly: 「Keep your existing SDK, change base_url and the key, no other code changes」 (据 relayrouter.io/docs 官方文档). Sending an Anthropic model ID to an OpenAI style endpoint, or the reverse, is a common cause of the error. Align the endpoint, base URL and model ID before retrying.

Resolution steps and billing note

Follow these steps to clear a "model not found" error and confirm billing impact.

1. Open relayrouter.io/models and copy the exact model ID.
2. Confirm the model belongs to a supported group (Claude, GPT-5.5, Gemini 3.5, DeepSeek, GLM, MiniMax, Moonshot).
3. Check that your base URL and endpoint (/v1/chat/completions or /v1/messages) match your SDK protocol.
4. Verify your API key is valid and change only the base_url and key per relayrouter.io/docs.
5. Resend the request with the corrected model ID.

Note on billing: failed or errored requests are generally not billed (source relayrouter.io), so a "model not found" response does not add a charge while you troubleshoot.

Comparison: two common causes

FAQ

Does a "model not found" error cost money? No. Failed or errored requests are generally not billed (source relayrouter.io).

Which protocols does RelayRouter support? Both OpenAI (/v1/chat/completions) and Anthropic (/v1/messages), per relayrouter.io/models.

Do I need to rewrite my code to fix model errors? No. 「Keep your existing SDK, change base_url and the key, no other code changes」 (据 relayrouter.io/docs 官方文档).