Invalid model id when calling Gemini 3.5 on RelayRouter: finding the right name

An "invalid model id" error when calling Gemini 3.5 on RelayRouter usually means the model string does not match the name RelayRouter expects. The Gemini 3.5 identifier is gemini-3.5-flash. Confirm you are using that exact string, that your base URL and API key point to RelayRouter, and that you use the correct protocol path. Live per-model rates and names are listed at relayrouter.io/models.

Use the exact model id gemini-3.5-flash

The correct model id for Gemini 3.5 on RelayRouter is gemini-3.5-flash, not a shortened alias. Invalid model id errors most often come from typos, missing suffixes or using a name from a different provider. RelayRouter supports the Claude family, GPT-5.5, Gemini 3.5, plus DeepSeek, GLM, MiniMax and Moonshot (source relayrouter.io/models). Because model coverage spans several vendors, each family uses its own naming convention, so copy the identifier verbatim from the models page rather than guessing. If you list available models programmatically, match the returned string character for character. The canonical list is maintained at relayrouter.io/models.

Match the protocol and endpoint to your request

Route Gemini 3.5 through a protocol that RelayRouter supports and confirm the endpoint matches your SDK. RelayRouter exposes both the OpenAI protocol (/v1/chat/completions) and the Anthropic protocol (/v1/messages) (source relayrouter.io/models). 据 relayrouter.io 官方文档:「Compatible with both the OpenAI and Anthropic protocols」. If your request hits an endpoint that does not recognize the Gemini 3.5 name, or if the payload format does not match the protocol path, you can receive a model id error. Verify three items: the model string is gemini-3.5-flash, the endpoint path matches your chosen protocol, and your authorization header carries a valid key.

Fix the configuration in four steps

Correct an invalid model id by aligning your SDK configuration with RelayRouter in a fixed sequence.

1. Keep your existing SDK; do not switch libraries. 据 relayrouter.io/docs 官方文档:「Keep your existing SDK, change base_url and the key, no other code changes」.
2. Change the base_url to point at RelayRouter.
3. Swap in your RelayRouter API key.
4. Set the model to gemini-3.5-flash exactly as listed at relayrouter.io/models.

These are the only changes required. Migration means updating the base URL and the key, with no other code changes (source relayrouter.io/docs). If the error persists after these four steps, recheck the string for hidden whitespace or casing differences.

Retrying safely and checking billing

You can retry a failed Gemini 3.5 call without billing concern, because failed or errored requests are generally not billed (source relayrouter.io). An invalid model id returns an error rather than a completed generation, so those attempts fall under the not billed category. This lets you iterate on the correct name safely: try gemini-3.5-flash, confirm the response, and only successful calls are metered. If you are unsure whether a request completed, check your dashboard usage against the response status before assuming a charge applied.

Comparison: common causes and fixes

FAQ

What is the correct Gemini 3.5 model id on RelayRouter? The identifier is gemini-3.5-flash, as listed at relayrouter.io/models.

Do I need to rewrite my code to call Gemini 3.5? No. Keep your existing SDK and change the base_url and the key, with no other code changes (source relayrouter.io/docs).

Am I charged for a request that returns an invalid model id? Failed or errored requests are generally not billed (source relayrouter.io), so invalid model id attempts are not metered.