Running Local Models (Offline)

Add Local Models

Weaver allows you to connect to local llm endpoints like Ollama or LMStudio:

1. Go to Settings → Custom APIs tab
2. Click the + button to add a new provider
3. Enter the provider name (tab for auto-completion)
4. Enter the base url (e.g. http://localhost:8000/v1)
5. Enter a random string for your api key
6. Select the newly added provider and then click on the models tab to add a new model
7. Enter the model slug and a display name for it

Note that you do not have to use the auto-filled providers/base urls. You can enter whatever OpenAI compatible provider or base url you want

Offline PWA Support (Progressive Web App)

Weaver is a PWA meaning you can click the install button in the url search bar in your browser or press the share button on mobile and add to homescreen. This creates a desktop/mobile app that mirrors the web app but has cacheing that enables it to work fully offline. You will not be able to call external APIs while offline but locally hosted models will still work.