Skip to main content

​
System Requirements

  • macOS: Sonoma (v14) or newer
  • Processor: Apple M series (CPU and GPU support) or Intel x86 (CPU only)

​
Installation

​
Quick Install

Install Ollama using the install script: 
curl -fsSL https://ollama.com/install.sh | sh

​
Manual Download

Alternatively, download Ollama.dmg and install:
1

Mount the DMG

Double-click the downloaded Ollama.dmg file.
2

Drag to Applications

Drag the Ollama application to your system-wide Applications folder.
Installing to the system Applications folder is recommended for automatic PATH configuration.
3

Launch Ollama

Open Ollama from your Applications folder.On first launch, Ollama will verify the CLI is in your PATH and prompt for permission to create a symlink in /usr/local/bin if needed.

​
Using Ollama

Once installed, Ollama runs in the background and the ollama command is available in Terminal.

​
Run Your First Model

ollama run gemma3

​
Check Running Models

ollama ps

​
View Installed Models

ollama list

​
Configuration

​
File Locations

Ollama stores files in the following locations:
LocationPurpose
~/.ollamaModels and configuration
~/.ollama/logsLog files
~/.ollama/logs/app.logGUI application logs
~/.ollama/logs/server.logServer logs
<install-location>/Ollama.app/Contents/Resources/ollamaCLI binary

​
Environment Variables

Set environment variables using launchctl: 
launchctl setenv OLLAMA_HOST "0.0.0.0:11434"
Then restart the Ollama application.

​
Common Environment Variables

VariableDescriptionDefault
OLLAMA_HOSTServer bind address127.0.0.1:11434
OLLAMA_MODELSModel storage location~/.ollama/models
OLLAMA_DEBUGEnable debug logging0
OLLAMA_NUM_PARALLELMax parallel requests1
OLLAMA_KEEP_ALIVEModel keep-alive duration5m

​
Changing Model Storage Location

If your home directory doesn’t have enough space for large models: 
launchctl setenv OLLAMA_MODELS "/Volumes/External/ollama-models"
Restart the Ollama application after changing the location.

​
Custom Installation Location

To install Ollama somewhere other than the system Applications folder:
1

Place the Application

Move Ollama.app to your desired location.
2

Add CLI to PATH

Ensure the CLI binary or a symlink is in your PATH:
sudo ln -s /your/custom/path/Ollama.app/Contents/Resources/ollama /usr/local/bin/ollama
3

Launch and Decline Move Prompt

On first start, decline the “Move to Applications?” request.

​
Apple Silicon (M-series) Benefits

Ollama is optimized for Apple Silicon:
  • Native ARM64: Built specifically for M1, M2, M3, and M4 chips
  • Unified Memory: Efficient GPU acceleration using shared memory architecture
  • Metal Support: Hardware-accelerated inference via Metal Performance Shaders
  • Neural Engine: Automatic optimization for Apple’s Neural Engine

​
Updates

Ollama automatically checks for updates. When an update is available:
  1. Click the Ollama icon in the menu bar
  2. Select “Restart to update”
Or download the latest version manually from ollama.com/download.

​
Disable Auto-Start on Login

If you prefer Ollama not to start automatically:
  1. Open System Settings
  2. Search for “Login Items”
  3. Find Ollama under “Allow in the Background”
  4. Click the toggle to disable
This setting persists across updates.

​
Logs and Debugging

​
View Application Logs

cat ~/.ollama/logs/app.log

​
View Server Logs

cat ~/.ollama/logs/server.log

​
Enable Debug Mode

launchctl setenv OLLAMA_DEBUG "1"
Restart the Ollama application, then check logs for detailed output.

​
Real-time Log Monitoring

tail -f ~/.ollama/logs/server.log

​
Troubleshooting

​
Ollama Command Not Found

If ollama is not recognized in Terminal:
  1. Check if the symlink exists: 
    ls -l /usr/local/bin/ollama
  2. If missing, create it manually: 
    sudo ln -s /Applications/Ollama.app/Contents/Resources/ollama /usr/local/bin/ollama
  3. Verify /usr/local/bin is in your PATH: 
    echo $PATH

​
Ollama Not Starting

  1. Check if the process is running: 
    ps aux | grep ollama
  2. Try launching from Terminal for error output: 
    /Applications/Ollama.app/Contents/MacOS/Ollama
  3. Check system logs: 
    log show --predicate 'process == "Ollama"' --last 5m

​
Insufficient Storage Space

Models can require tens to hundreds of GB:
  1. Check available space: 
    df -h ~
  2. Change model storage location: 
    launchctl setenv OLLAMA_MODELS "/Volumes/External/models"
  3. Restart Ollama

​
Performance Issues

  • Close other applications: Free up RAM for model inference
  • Check Activity Monitor: Look for high CPU/memory usage
  • Try smaller models: Start with models like gemma2:2b or llama3.2:1b
  • Update macOS: Ensure you’re running the latest version for best Metal performance

​
Uninstallation

To completely remove Ollama from your Mac: 
sudo rm -rf /Applications/Ollama.app
sudo rm /usr/local/bin/ollama
rm -rf ~/Library/Application\ Support/Ollama
rm -rf ~/Library/Saved\ Application\ State/com.electron.ollama.savedState
rm -rf ~/Library/Caches/com.electron.ollama
rm -rf ~/Library/Caches/ollama
rm -rf ~/Library/WebKit/com.electron.ollama
rm -rf ~/.ollama

​
API Access

Ollama’s REST API is available at http://localhost:11434: 
curl http://localhost:11434/api/generate -d '{
  "model": "gemma3",
  "prompt": "Why is the sky blue?",
  "stream": false
}'
See the API documentation for complete reference.

​
Next Steps

Quickstart Guide

Get started with your first model

Model Library

Browse available models

API Reference

Integrate Ollama into your apps

CLI Reference

Master the command-line interface

Build docs developers (and LLMs) love

Get started for freeTalk to us