Cline setup with UniRoute
Configure Cline in VS Code with UniRoute’s OpenAI Compatible endpoint, API key, and model, then verify requests and fix common errors.
Cline · pay-as-you-go API
Connect Cline’s OpenAI Compatible provider to UniRoute with an API address, key, and model. Use the configuration for chat and coding tasks in VS Code.
Best for: Developers planning, writing, and editing code in VS Code.
Quick setup
Sign in to create a key and fill it in automatically Sign in
Open Cline → Settings (gear) → API Configuration and copy each value into its matching field.
API Provider
OpenAI CompatibleBase URL
https://api.uniroute.app/v1OpenAI Compatible API Key
YOUR_UNIROUTE_API_KEYModel ID
deepseek-v4-flashSetup steps
1. Open the local provider settings
Click the gear in the Cline sidebar and open API Configuration. If the Local / Cloud selector is shown, choose Local.
This configures Local; Cloud tasks have separate model settings.
2. Fill in the endpoint and key
Set API Provider to OpenAI Compatible, then fill Base URL and OpenAI Compatible API Key with the values above. Leave Custom Headers empty, Use Azure Identity Authentication unchecked, and Set Azure API version disabled.
3. Select the model
Choose deepseek-v4-flash in Model ID. For manual entry, select Use custom model ID… and enter the same ID in Custom Model ID. Keep /v1 in Base URL; Cline adds /chat/completions.
4. Save the Plan and Act settings
For one shared configuration, turn off Use different models for Plan and Act modes. For separate configurations, set the provider and model in both Plan and Act, then save settings.
For the initial text task, keep context, output, image, and sampling settings under Model Configuration at their defaults.
What these fields do
| Field | Purpose |
|---|---|
| API Provider | Selects the Chat Completions request format. |
| Base URL | API prefix; Cline appends /chat/completions or /models. |
| OpenAI Compatible API Key | Authenticates requests to UniRoute. |
| Model ID / Custom Model ID | The complete model ID sent to the API. |
| Model Configuration | Model capability metadata and optional request parameters; keep defaults for initial text setup. |
Choose a model
Confirm the connection
Send a short test: Reply only: OK.
Open usage logs to review the model, request status and charge. The test is a real, billed API request.
After the reply, ask Cline to read one small workspace file and name it. Approve the file read when prompted, then find the follow-up request in usage logs.
Switch back / restore a backup
Return to API Configuration, select your previous provider and model, and save. Restore both modes if Plan and Act have separate configurations.
Not installed yet?
Install Cline from the VS Code Extensions panel and open its sidebar. See the official Cline installation guide.
FAQ
Do I need ClinePass?
Select OpenAI Compatible and use your UniRoute key. Requests use UniRoute’s usage-based billing.
What if the model list is empty?
Select Use custom model ID…, enter the complete ID in Custom Model ID, then check the Base URL and key.
Should Base URL include /chat/completions?
Use https://api.uniroute.app/v1. Cline adds the endpoint path.
Troubleshooting
401 / Invalid API Key
The key is incomplete, revoked, or unsaved.
Paste the complete valid key and save settings.
404
Base URL contains a repeated endpoint path.
Use https://api.uniroute.app/v1 without /chat/completions.
Model Not Found
The model ID or key permissions need attention.
Enter the complete model ID and check key permissions and wallet funds in the console.
Act uses a different model
Plan and Act have different saved configurations.
Configure both modes or turn off Use different models for Plan and Act modes.
Related links
Manage API keys
Model prices