How to Install Ollama on Mac — Complete Setup Guide (2026)

Step 1: Download Ollama

Head to ollama.com and click the download button for macOS. The installer is a standard .dmg file, around 90 MB. It supports every Apple Silicon Mac — M1, M2, M3, M4, and M5 — as well as Intel Macs (though performance will be significantly slower on Intel).

Once the download finishes, open the .dmg file and drag the Ollama icon into your Applications folder, just like any other Mac app.

Step 2: Install and Verify

Launch Ollama from your Applications folder. On first launch, it will ask for permission to install its command-line tool. Click Allow — this adds the ollama command to your Terminal so you can interact with it from the command line.

Open Terminal (search for it in Spotlight with Cmd + Space) and verify the installation:

ollama --version

You should see something like ollama version 0.6.x. If you see this, Ollama is installed and ready to go.

Step 3: Pull Your First Model

Ollama uses a Docker-like pull system to download models. For your first model, we recommend Qwen 3.5 9B — it is one of the best small models available in 2026, with strong reasoning and coding ability.

ollama pull qwen3.5:9b

This downloads the Q4_K_M quantized version (about 5.5 GB). The download speed depends on your internet connection — on a typical broadband connection, expect 2-5 minutes.

If you have an 8 GB Mac and want something even lighter, try:

ollama pull qwen3.5:4b

For 32 GB+ Macs looking for maximum capability:

ollama pull qwen3.5:32b

Step 4: Start Chatting

Run the model with a single command:

ollama run qwen3.5:9b

You will see a prompt appear in your terminal. Type any question or task and press Enter. The model runs entirely on your Mac — nothing is sent to the internet.

Try these to get started:

• "Explain quantum computing in simple terms" — tests general knowledge
• "Write a Python function to reverse a linked list" — tests coding ability
• "Summarize the key points of this text: [paste text]" — tests comprehension

To exit the chat, type /bye or press Ctrl+D.

Step 5: Essential Commands Cheat Sheet

Here are the commands you will use most often with Ollama:

# List all downloaded models
ollama list

# Pull a specific model
ollama pull llama4-scout

# Run a model (downloads it first if needed)
ollama run mistral

# Remove a model to free disk space
ollama rm qwen3.5:9b

# Show model details (size, parameters, quantization)
ollama show qwen3.5:9b

# Start the API server (for use with other apps)
ollama serve

# Check running models
ollama ps

Connecting Other Apps to Ollama

Ollama works as a backend for many popular AI apps. Once Ollama is running, you can connect:

• LM Studio — a beautiful desktop GUI that can use Ollama as its backend
• Jan — an open-source ChatGPT alternative that connects to local Ollama models
• Open WebUI — a web-based ChatGPT-like interface that works with Ollama out of the box
• Continue.dev — a VS Code extension for AI-assisted coding using local models

For a complete list of compatible apps, check our software directory. To see how your Mac performs with different models, visit the leaderboard.