Local AI › Local AI on a Mac with Apple Silicon: guide and models

Local AI on a Mac with Apple Silicon: guide and models

Macs with Apple Silicon run local AI with the same tools as a PC, and some offer versions optimised for the Mac (MLX format).

Checked on 13 Oct 2026. Versions change every week: check the official site before installing.

Which tools work

• LM Studio: macOS 14.0 or later on Apple Silicon, 16 GB of RAM or more recommended (requirements). It can download models in MLX format.
• Ollama: app for macOS (ollama.com/download). Its library has MLX variants, e.g. gemma4:e4b-mlx and qwen3.5:2b-mlx.
• Jan and GPT4All: macOS installers (.dmg) listed in their READMEs.

Memory on a Mac

On Apple Silicon the memory is unified: the CPU and the GPU share it, so what matters is your Mac's total memory rather than a separate graphics card. macOS keeps part of it for itself; we do not give an exact usable figure because we could not verify it in an official source. As a guide, use the table of models by memory and leave some margin.

Install step by step

How to install local AI on a Mac, step by step.

Keep reading

• Local AI: how to run AI on your computer offline (2026)
• Tools to run AI locally: Ollama, LM Studio, Jan and more
• How to install local AI on Windows, step by step
• How to install local AI on a Mac, step by step
• How to install local AI on Linux, step by step
• Which local AI model to pick for your RAM: 8, 16, 24, 32 GB
• Generate AI images locally with ComfyUI
• Local AI vs. cloud AI: privacy, cost and limits
• Local AI: frequently asked questions