Chatbox API setup
Connect Chatbox to UniRoute with an OpenAI-compatible provider, API Host, API Path, key and model.
Chatbox · pay-as-you-go API
Add a UniRoute provider using OpenAI API Compatible. Enter the host and request path separately, add a model, then start a conversation.
Best for: People using Chatbox for everyday conversations, writing and code questions.
Quick setup
Sign in to create a key and fill it in automatically Sign in
Copy each value into its matching client field.
API Mode
OpenAI API CompatibleAPI Host
https://api.uniroute.app/v1API Key
YOUR_UNIROUTE_API_KEYAPI Path
/chat/completionsName
UniRouteModel ID
deepseek-v4-flashSetup steps
1. Add a compatible provider
Open Settings → Model Provider → Add. Set Name to UniRoute, choose OpenAI API Compatible in API Mode, and click Add.
2. Fill the host, path and key
Paste API Key, API Host and API Path from Quick setup. Check Preview: the address is https://api.uniroute.app/v1/chat/completions. Keep Improve Network Compatibility off for the initial connection.
3. Add a chat model
Under Model, click New, enter deepseek-v4-flash in Model ID, and save. Alternatively, click Fetch and select models from the returned list.
Set model capabilities for the features you use; leaving capabilities unconfigured treats the model as text-only.
4. Check and select the model
Click Check, return to the conversation screen, and select the model under UniRoute.
Chatbox AI hosted models and search are configured separately from your UniRoute provider.
What these fields do
| Field | Purpose |
|---|---|
| API Mode | OpenAI API Compatible selects Chat Completions. |
| API Host | The server address, including /v1. |
| API Path | The /chat/completions path joined to API Host; keep /v1 in the host only. |
| API Key | Your UniRoute Bearer credential. |
| Model ID | The exact model identifier used in requests. |
| Improve Network Compatibility | An optional connection setting; leave it off for the initial setup. |
Choose a model
Confirm the connection
Send a short test: Reply only: hello
Open usage logs to review the model, request status and charge. The test is a real, billed API request.
Click Check, then open a new conversation with your UniRoute model. Read the streamed reply and match the model in usage logs.
Switch back / restore a backup
Select your previous provider’s model in the conversation. Remove only the UniRoute provider from Model Provider if you no longer need the connection.
Not installed yet?
Install Chatbox from the official download page, then open Settings → Model Provider.
FAQ
Does API Path include /v1?
Use /chat/completions when API Host is https://api.uniroute.app/v1. Preview shows the complete URL.
Can Chatbox fetch the model list?
Yes. Fetch requests the configured host’s /models endpoint with your key. Select the returned models to add them.
Can I use Claude through this provider?
Yes. Select the Claude Sonnet model shown above and keep OpenAI API Compatible; UniRoute supports Chat Completions for that model.
Can this connection generate images?
The custom OpenAI adapter supplies text-chat models and does not add an image-generation model.
Troubleshooting
404 / repeated /v1
The host and path contain a repeated version or endpoint.
Use API Host https://api.uniroute.app/v1 and API Path /chat/completions.
401 / invalid key
The key is incomplete, revoked or lacks permission.
Paste a complete active UniRoute key into API Key.
Model missing from the selector
The provider’s model list is empty.
Add a Model ID with New, or select models returned by Fetch.
Chatbox AI search asks for a license
The selected search service has its own license.
Choose and configure your search provider separately in Chatbox.
References
Chatbox model configuration · Bring your own key · Host and path handling
Related links
All integrations · Manage API keys · Protocols and addresses · Model prices · Usage records