AI Code With LogoAI Code With
IntegrationsRegionsLogin

Chatbox

How to fix Chatbox reporting Model Not Found or an unknown model

Chatbox does not sync the AI Code With catalog automatically. Fetch may return an empty list, so chat can only use Model IDs you fetch or type. The ID must match the AI Code With docs exactly; Gemini models also need Google Gemini API compatible mode on that model.

Start with these checks

The requested Model ID does not match a model AI Code With offers, it was mistyped, or Gemini is still on OpenAI-compatible mode.

  • The request reports Model Not Found, unknown model, or 404, or the typed model is missing from the chat page.
  • The Model ID was typed by hand, has extra spaces, or used the display name instead of the ID
  • Copy the exact Model ID from this page (e.g. deepseek-v4-flash)
model not foundunknown model404

01

Understand how Chatbox uses the Model ID

The provider name can be anything, but the ID sent upstream must be a real AI Code With ID such as deepseek-v4-flash, not “DeepSeek V4 Flash”.

If the list does not show the new model, Fetch probably failed and you did not click New, or the chat page is still on a built-in provider. Confirm you are using the custom provider you added.

02

Re-enter the exact ID from the integration page

Copy the target ID from the model list on this page and paste it with New in Chatbox. Do not type from memory, and do not add a provider prefix (openai/deepseek-v4-flash is wrong).

Model Not Found is almost always the ID itself or Gemini mode. A wrong host usually shows up as 404 or a timeout — use the API host article first.

  1. 1Copy the exact Model ID from this page (e.g. deepseek-v4-flash)
  2. 2In Chatbox, click New and paste it; do not type it, change casing, or add an openai/ prefix

03

Check Gemini mode, access, and select the model on the chat page

Gemini models need Google Gemini API compatible mode in that model's settings; leave other models on OpenAI API compatible. If the ID is correct but the error remains, confirm the model is currently available.

After saving, pick that entry on the chat page before sending. If you add the model but leave another model selected, the request never uses the new ID. When replies look normal, match the model name in usage records.

  1. 1In Chatbox, click New and paste it; do not type it, change casing, or add an openai/ prefix
  2. 2For Gemini models, open that model's settings and switch API mode to Google Gemini API compatible

Chatbox does not sync the AI Code With catalog automatically. Each new model requires fetching or creating the Model ID again.

If it is still not fixed

401 / Invalid API KeyFetched model list is empty

Sources

Reconfigure

Return to the integration center and continue the setup steps.

Return to the related setup step