Skip to main content
AI Chat Importer
Desktop App · Setup Guide

How to Set Up Ollama for Auto-Sort

A complete beginner guide to installing Ollama, choosing the right AI model for your hardware, and using Auto-Sort to automatically organise your imported conversations — all on your own machine.

What is Ollama?

Ollama is a free, open-source tool that lets you run AI language models directly on your own computer. Instead of sending your text to a cloud service like ChatGPT, everything happens locally — the AI runs on your machine, processes your data, and never sends anything over the internet.

AI Chat Importer's Auto-Sort feature uses Ollama to read your imported conversations and suggest a folder structure for them. Because Ollama runs locally, your conversation content stays completely private — nothing leaves your device.

Can my computer run it?

Ollama runs on most modern computers. Select your hardware below to get a personalised model recommendation.

Not sure what GPU you have? On Windows, press Win + R, type dxdiag, and press Enter — look under the Display tab. On Mac, click the Apple menu → “About This Mac”.

Step 1

Install Ollama

  1. Go to ollama.com and download the installer for your operating system (Windows, Mac, or Linux).
  2. Run the installer and follow the on-screen steps.
  3. Once installed, Ollama runs in the background automatically — you don't need to launch it manually each time.
Windows users: After installing, you'll see the Ollama icon in your system tray (bottom-right of the taskbar). This means it's running.
Step 2

Download the AI model

Ollama itself is just the runner — you also need to download an AI model for it to use. Use the hardware checker above to find the right model for your machine. The most common recommendation is llama3.1:8b.

  1. Open a terminal:
    • Windows: press Win + S and search for “Command Prompt” or “PowerShell”
    • Mac: press Cmd + Space and search for “Terminal”
  2. Run the following command (replace llama3.1:8b with your recommended model if different):
    ollama pull llama3.1:8b
  3. This downloads the model file — it's around 4.7 GB so may take a few minutes depending on your connection speed.
  4. Once complete, run ollama list to confirm the model is installed. You should see it listed with its name and size.
Step 3

Configure AI Chat Importer

  1. Open the AI Chat Importer desktop app.
  2. Click the settings cog icon in the sidebar.
  3. Select “Ollama Settings”.
  4. Set the URL to http://localhost:11434 — this is the default and should work unless Ollama is running on a different machine on your network.
  5. Set the model name to llama3.1:8b (or whatever model you installed — it must match exactly what ollama list shows).
  6. Click “Test Connection” to confirm everything is working — you should see a green “Connected” status.
Step 4

Run Auto-Sort

  1. Import your conversations if you haven't already — use the import button in the sidebar.
  2. Click the settings cog in the sidebar and select “Auto-Sort Conversations”.
  3. AI Chat Importer will scan your conversations and suggest a folder structure based on topics and themes.
  4. Review the suggested folders and make any changes you'd like, then confirm — your conversations will be sorted automatically.
Auto-Sort learns from your corrections. If you move a conversation to a different folder after sorting, it will take that into account next time.

Troubleshooting

“Ollama not found” or connection refused

Make sure Ollama is running. On Windows, look for the Ollama icon in the system tray (bottom-right of the taskbar). On Mac, check the menu bar at the top of the screen. If it's not running, open the Ollama app to start it, then try again.

“Model not found”

The model name in Ollama Settings doesn't match what 's installed. Run ollama list in your terminal to see the exact model name, and copy it precisely into the settings field — spelling and casing must match exactly.

Auto-Sort is very slow

This is normal on CPU-only machines — each conversation takes a few seconds to process, so a batch of 50 conversations may take several minutes. Consider switching to llama3.2:3b for faster results (though slightly less accurate). Run ollama pull llama3.2:3b to download it.

Ollama is on a different computer on my network

Change the URL in Ollama Settings from http://localhost:11434 to http://[your-machine-ip]:11434 — for example, http://192.168.1.69:11434. You can find your machine's local IP address in your network settings.

You're all set

That's everything you need to get Auto-Sort running. Once it's configured, sorting hundreds of conversations takes just a few minutes — and because everything runs locally, your chat history stays completely private.

Haven't downloaded AI Chat Importer yet? Get the desktop app →