What Is Ollama?

Ollama is an open-source tool that lets you download and run large language models locally on your computer — no cloud, no API keys, no data leaving your machine.

Why Use Ollama?

• Privacy: Your conversations never leave your Mac
• No API costs: Run unlimited queries at zero cost
• Offline access: Works without an internet connection
• Speed: Local inference on Apple Silicon is surprisingly fast

Getting Started

# Install Ollama
brew install ollama

# Download and run a model
ollama run qwen3

# List available models
ollama list

Popular Models on Ollama

Model	Best For
Qwen3.8 27B	Best all-round on a 24GB GPU or 32GB Mac
Llama 3.2 3B	Lightweight tasks, widest tool support
DeepSeek-R1	Reasoning and chain-of-thought
Gemma 3 4B	Fast, efficient responses
Devstral Small 24B	Code generation
Phi-3 Mini	Very small hardware footprint

Ollama + Elvean

Elvean connects to Ollama natively — just point it at your local Ollama server and all your downloaded models appear automatically. Switch between local and cloud models mid-conversation with @mentions.