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

A "model not found" error on RelayRouter usually means the model ID in your request does not match a model your API key can access, or it belongs to a group your key is not authorized for. RelayRouter supports the Claude line, 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, then verify your key's access before retrying.

What causes a model not found error

A model not found error means the requested model ID is not recognized or not enabled for your key. RelayRouter is protocol compatible on two fronts: it accepts OpenAI requests at /v1/chat/completions and Anthropic requests at /v1/messages. 据 relayrouter.io 官方文档,「同时兼容 OpenAI 与 Anthropic 两种协议」. If you send a model name that does not exist in the supported set (the Claude series, GPT-5.5, Gemini 3.5, DeepSeek, GLM, MiniMax, Moonshot), the gateway cannot route the call. Typos, deprecated aliases and models outside your key's group are common triggers. Check the exact spelling against relayrouter.io/models, which lists the current model range and per-model rates.

How to check your key's model group

Verify your key's model group by matching the model ID to the supported list, then confirming your endpoint and key are correct.

1. Open relayrouter.io/models and locate the exact model ID you intend to call.
2. Confirm the model belongs to a supported family: Claude series, GPT-5.5, Gemini 3.5, DeepSeek, GLM, MiniMax or Moonshot.
3. Check that your request uses the matching protocol endpoint: /v1/chat/completions for OpenAI or /v1/messages for Anthropic.
4. Verify your base_url and API key point to RelayRouter.
5. Retry the request. Failed or errored requests are typically not billed, so retries carry no charge for the failed attempt.

Fixing the base_url and key without code changes

Most model routing issues resolve by correcting the base URL and key alone. Migration requires no rewrite of your application logic. 据 relayrouter.io/docs 官方文档,「保留现有 SDK,改 base_url 与 key 即可,无需改其他代码」. If your requests fail because the client is still pointed at an upstream provider, the model ID may be valid there but unrecognized by RelayRouter's group mapping. Update the base_url to the RelayRouter endpoint and swap in your RelayRouter key, then keep the existing SDK. Two protocols are covered (OpenAI and Anthropic), so most SDKs work unchanged once the endpoint and credentials are set. Because failed or errored requests are typically not billed, you can iterate on configuration without accumulating charges.

Endpoint and protocol reference

The correct endpoint depends on which protocol your SDK speaks.

RelayRouter supports both protocols simultaneously, so a valid model ID sent to the wrong endpoint can still fail. For example, a Claude series model called through the Anthropic path (/v1/messages) and a GPT-5.5 request through the OpenAI path (/v1/chat/completions) both route correctly only when the model ID is confirmed against relayrouter.io/models. See the migration notes at relayrouter.io/docs.

FAQ

Which models can I call on RelayRouter? The supported range includes the Claude series, GPT-5.5, Gemini 3.5, plus DeepSeek, GLM, MiniMax and Moonshot. The current list and rates are at relayrouter.io/models.

Do I need to rewrite my code to fix the error? No. Keep your existing SDK and change only the base_url and key, as noted at relayrouter.io/docs.

Am I charged if the model not found error occurs? Failed or errored requests are typically not billed, so retrying with a corrected model ID does not incur charges for the failed attempt.