Skip to content

Local AI Setup

DocGraph’s AI features run entirely on your machine using Ollama. Nothing is sent to the cloud, and AI is off by default — you opt in when you’re ready.

  1. Install Ollama from ollama.com and make sure it’s running.
  2. In DocGraph, open the AI setup wizard (it appears automatically on first run, or find it in settings).
  3. Choose a chat model and an embedding model; DocGraph will guide you through pulling them.
  4. Enable AI. DocGraph indexes your vault locally so the assistant can answer questions about your notes.

Local models need a reasonably modern machine — an Apple Silicon Mac or a PC with a decent GPU gives the best experience. Smaller models run fine on modest hardware for basic tasks.

  • Your notes are embedded and indexed locally.
  • Chat requests go to your local Ollama instance only.
  • Disabling AI stops all indexing and model usage.