Log inCreate wallet

Run an LLM on your Mac.

Apple Silicon Macs are some of the best machines for running language models locally: private, offline and fast. Here is how to start in five minutes, and how to pick a model that fits.

Updated

The short version

  • Any Mac with Apple Silicon can run a local LLM. Memory decides which models fit; memory bandwidth decides how fast they run.
  • The quickest way to start: LM Studio (an app) or Ollama (one command). For the most speed and control: Apple's MLX with mlx-lm.
  • Rule of thumb: a 4-bit model needs a little over half a gigabyte of memory per billion parameters, plus room for the context.
  • Measure your Mac on MLX Bench and compare it with every other Apple Silicon chip.

1What fits on your Mac

A model's weights have to fit in memory with room to spare for the context and macOS. At 4 bits per weight, plan on about 0.55 to 0.6 GB per billion parameters.

Mac memoryComfortable model size (4-bit)Examples of the class
8 GBup to about 4Bsmall chat and coding models
16 GBup to about 14Bmost popular 7B–14B models
32 GBup to about 30B, or mixture-of-experts models of similar sizemid-size reasoning and coding models
64 GBup to about 70Blarge dense models
128 GBup to about 120B (mixture-of-experts)gpt-oss-120b, about 65 GB
512 GBhundreds of billions of parametersQwen3.5-397B-A17B, about 220 GB
Approximate; quantization, context length and other apps change the numbers. Parameter counts from the model cards (gpt-oss-120b, Qwen3.5-397B-A17B).

Speed is a separate question: each new token needs the active weights read from memory once, so tokens per second scale with memory bandwidth. Our guide to the best Mac for local LLMs covers every chip.

2The easy way: LM Studio

LM Studio is a free Mac app with a chat window and a model browser. On Apple Silicon it can run models in Apple's MLX format as well as the GGUF format used by llama.cpp.

  1. Download LM Studio and open it.
  2. Search for a model, prefer an MLX version on Apple Silicon, and pick a size that fits your memory.
  3. Load it and start chatting. You can also turn on its local server for other apps.

3One command: Ollama

Ollama runs models from the terminal and serves them to other apps on your Mac. After installing it, one command downloads and starts a model:

ollama run llama3.2

Since version 0.19, Ollama on Apple Silicon has a preview powered by Apple's MLX framework for supported models, with large speedups; the preview asks for a Mac with more than 32 GB of memory (Ollama).

4The fastest way: Apple's MLX and mlx-lm

MLX is Apple's open-source machine learning framework for Apple Silicon, and mlx-lm is its package for language models. Thousands of models are ready to use from the mlx-community organization on Hugging Face.

pip install mlx-lm
mlx_lm.generate --model mlx-community/Llama-3.2-3B-Instruct-4bit --prompt "Explain unified memory in one paragraph"

Use mlx_lm.chat for a conversation in the terminal and mlx_lm.server for a local API that other apps can call. On M5 chips, MLX uses the new neural accelerators in the GPU: Apple measured the first token arriving 3.3 to 4.1 times sooner than on M4 (Apple Machine Learning Research).

5Measure it, then put it to work

MLX Bench is an open leaderboard of MLX speed on every Apple Silicon chip: generation and prompt tokens per second and peak memory. Run the one-file script on your Mac and see where it lands.

When you're not using your Mac, it can run AI for others. MLX Node is a free app that runs AI jobs for MLX Network on the same MLX framework and earns MLX Tokens for verified work, within the limits you set.

6Questions

Can a MacBook Air run an LLM?
Yes. A MacBook Air with 16 GB runs 7B to 14B models at 4-bit comfortably; with 32 GB, models up to about 30B. It will be slower than a Mac with a Pro or Max chip, which has more memory bandwidth.
Is MLX faster than Ollama on a Mac?
Often, because MLX is written for Apple Silicon. Ollama's own MLX-powered preview reports large speedups on Apple Silicon. Our MLX vs Ollama guide compares them in detail.
Do local models work offline?
Yes. Once downloaded, the model runs entirely on your Mac and your prompts never leave it.
How much memory do I need for a 70B model?
About 40 GB for the weights at 4-bit, so a Mac with 64 GB or more. Mixture-of-experts models with similar quality may need more memory but run faster.

Sources

  1. gpt-oss-120bhuggingface.co
  2. Qwen3.5-397B-A17Bhuggingface.co
  3. LM Studiolmstudio.ai
  4. Ollamaollama.com
  5. Ollamaollama.com
  6. MLXgithub.com
  7. Apple Machine Learning Researchmachinelearning.apple.com

Read next

Run AI on Apple SiliconThe best Mac for local LLMs.Which Mac should you buy to run LLMs locally? Memory decides what fits and bandwidth decides speed. Every Apple Silicon chip compared, from MacBook Air to Mac Studio M5 Ultra.Read the guideRun AI on Apple SiliconMLX vs Ollama vs llama.cpp.Apple's MLX, Ollama and llama.cpp compared for running LLMs on Apple Silicon: speed, model formats, ease of use and when to pick each. Updated for Ollama's MLX preview.Read the guideRun AI on Apple SiliconRun DeepSeek V4.1 Flash on a Mac.How to run DeepSeek V4.1 Flash (and 1017) locally on a Mac: how much RAM each MLX and GGUF build needs, which Mac Studio holds it, real tokens per second and commands.Read the guide
How to run an LLM locally on your Mac (MLX, Ollama, LM Studio)