Ollama

• https://github.com/ollama/ollama
• It's a platform for running LLMs locally on your machine
• It's an independent open-source project (not by meta, openai, google, etc)

Popular models supported

• Library of supported models: https://ollama.com/library/gpt-oss

• LLaMA (Meta)

• https://ollama.com/library/llama3.2
• llama3.2:latest (2.0GB and supports tools API)
• llama2-uncensored:latest (3.8 GB)
• Mistral (Mistral AI)
• Gemma (Google DeepMind)
• https://ollama.com/library/gemma3
• gemma3:270m: (300MB very light but do not support tools API)
• Qwen (Alibaba)
• Phi series (Microsoft)
• GPT OSS (OpenAI)
• https://ollama.com/library/gpt-oss
• gpt-oss:20b (14 GB)
• gpt-oss:120b (65 GB)

Install

brew install ollama-app --cask