Custom Endpoint
Connect any OpenAI-compatible API to Sciwand via the Custom Endpoint option.
Use Custom Endpoint to connect any OpenAI-compatible API - a self-hosted gateway, a proxy, or a provider not listed by name.
Cost and subscription depend entirely on the endpoint you point to. A self-hosted/local endpoint is free; a third-party gateway may be pay-as-you-go.
Step 1 - at the endpoint
- Have your server running and note its base URL (for example
http://localhost:1234orhttps://api.example.com). - If it requires authentication, have the API key/token ready.
Step 2 - in Sciwand
- Open Settings -> AI Models, click Manage (or Add Model), then select Custom Endpoint and click Configure.
- Set Server URL. Sciwand auto-detects an OpenAI-style
/v1path. - If the endpoint requires authentication, fill in API Key (Optional). Leave it blank for open local servers.
- Click Save Changes, then tick the models Sciwand discovers.
Notes
- The endpoint must implement the OpenAI chat-completions API (
/v1/chat/completions) or the Ollama tags API. - This option also exists on the Embeddings tab - see Local models.