Skip to content
OpenWaggle

Ollama (Local Models)

Running AI models locally with Ollama — no API key or internet connection required.

Coming soon — Ollama support is implemented but not yet fully tested. It may not work reliably in the current release.

Disabled by default. Ollama runs models locally on your machine — no API key or internet connection required.

Setup

  1. Install Ollama on your machine.
  2. Pull a model: ollama pull llama3.2 (or any model you prefer).
  3. In OpenWaggle Settings > Connections, enable Ollama.
  4. The default base URL is http://localhost:11434. Change it if your Ollama instance runs elsewhere.

Dynamic Model Discovery

OpenWaggle automatically detects all models installed in your Ollama instance. The model list refreshes when you open the model picker.

Custom Base URL

Point OpenWaggle to a remote Ollama instance by changing the base URL in settings (e.g., http://192.168.1.100:11434).