Connect an OpenAI-Compatible Service

Use OpenAI Compatible for OpenAI and for any provider that exposes an OpenAI-compatible chat API.

Add the service

1. Open Settings and go to Services.
2. Add a new service or edit an existing one.
3. Set API Format to OpenAI.
4. Enter a Name you will recognize.
5. Keep the pre-filled Service Address for OpenAI, or replace it with your provider's address.
6. Choose Address Type.
7. Enter and save the Access Key.
8. Add the Model ID your provider expects.
9. Click Test Connection.

Open Settings in SpotAsk

Opens the Service page in SpotAsk. If SpotAsk is not installed this link does nothing — open it by hand instead: SpotAsk menu bar icon → Settings... → Service, or press ⌘ + , to open Settings. Direct link: spotask://settings/provider

Address examples

Use Service Root when the provider gives you a base address:

https://api.openai.com/v1

Use Full Request Address when the provider gives you the complete chat endpoint:

https://api.openai.com/v1/chat/completions

Other providers may use different paths. Copy the exact base address or full endpoint from your provider's documentation.

Get a key for OpenAI

SpotAsk pre-fills https://api.openai.com/v1. Create a key at platform.openai.com/api-keys, paste it into Access Key, and click Test Connection. The request and response fields are documented at platform.openai.com/docs/api-reference.

Add the model

The model list can be refreshed when the service uses Service Root and exposes a models endpoint. If discovery is not available, add the model manually:

1. Under the service, select Add Model.
2. Enter a Display Name for the model picker.
3. Enter the exact Model ID the API expects.
4. Choose whether answers should stream into the window.
5. Select Use for Chat when ready.

After setup

Close Settings, press ⌥ + Space, and ask a question. If the connection test passes but a question fails, see Troubleshooting.

Related: Service Root vs Full Request Address, Providers & Models