Chatbox
How to fix an empty Chatbox model list after Fetch
Chatbox Fetch depends on the upstream returning a model list. AI Code With may not return a full catalog, so an empty list can always be fixed by clicking New and typing the Model ID. Chat does not depend on the fetched list — an ID is enough.
Start with these checks
Fetch returned no models. Recheck the host and key, then create the full Model ID by hand.
- Fetch returns an empty list, or chat cannot pick the AI Code With model you just configured.
- The API host or key is wrong, so the fetch request fails
- Confirm the API host is https://api.aicodewith.ai (no /v1) and the key is pasted correctly
01
Recheck the host and key first
Fetch often fails because the host includes /v1 or the API key is invalid. Confirm the host is https://api.aicodewith.ai, the path is /chat/completions, then re-paste the key.
If Check already returns 401, follow the invalid API key article first instead of repeatedly clicking Fetch.
02
Create the model by hand if Fetch is empty
Click New and paste the exact Model ID from the integration page. Do not enter the display name, and do not add an openai/ prefix.
For Gemini models, after creating the ID switch that model's API mode to Google Gemini API compatible.
- 1Confirm the API host is https://api.aicodewith.ai (no /v1) and the key is pasted correctly
- 2Click Fetch again
03
Save and verify in chat
Save the provider, pick that custom provider and the new model on the chat page, and send a test message.
A normal reply plus a matching usage-record entry means the setup works. A successful Fetch is not a successful connection.
- 1Click Fetch again
- 2If the list is still empty, click New and paste the full Model ID (e.g. deepseek-v4-flash)
An empty Fetch is not a failed setup. Creating the full Model ID by hand is a normal Chatbox path to AI Code With.
If it is still not fixed
Sources
Reconfigure
Return to the integration center and continue the setup steps.
Return to the related setup step