1. Install Ollama Server (Optional)
If you want to use Ollama models, ReactorAI relies on the Ollama server to run AI models. You can install it locally or use a remote server.
- Visit ollama.com to download the server
- Follow platform-specific setup instructions
- Run a test command:
ollama run llama2to verify it works
1a. How to Install Ollama and Download Models
Installing Ollama
macOS & Linux:
curl -fsSL https://ollama.com/install.sh | sh
Windows:
Download the installer from ollama.com/download/windows and run it.
Downloading Popular Models
Essential Models for ReactorAI:
# General purpose model (recommended for beginners)
ollama pull llama3.2
# Best for coding tasks
ollama pull codellama
# Fast, lightweight model
ollama pull phi3
# Great for multilingual tasks
ollama pull mistral
# Advanced reasoning (larger model)
ollama pull llama3.1:70b
Essential Commands
# List installed models
ollama list
# Start ollama service
ollama serve
# Test a model interactively
ollama run llama3.2
# Remove a model
ollama rm modelname
# Pull specific model version
ollama pull llama3.2:3b
Tips for ReactorAI Users
- Start Small: Begin with
llama3.2orphi3- they're fast and work great - Check System Resources: Larger models (70b+) need 32GB+ RAM
- Keep Ollama Running: ReactorAI connects to the ollama service on
http://localhost:11434 - Model Switching: Download multiple models and switch between them in ReactorAI
1b. Setup Gemini API (Alternative)
If you prefer to use Google's Gemini models instead of or alongside Ollama:
- Visit Google AI Studio
- Create an API key for Gemini
- Keep the API key ready for app configuration
2. Install ReactorAI on macOS
- Go to the Mac App Store
- Click “Get” to install
- Launch the app and configure the Ollama server address or Gemini API key
3. Install ReactorAI on iOS
- Open the App Store on your iPhone or iPad
- Search for "ReactorAI" and install the app
- Open the app and set your Ollama server (local or remote) or Gemini API key
4. Install ReactorAI on Windows
- Go to the Microsoft Store
- Click “Install” to get ReactorAI
- Open the app and provide the Ollama model server URL or Gemini API key