Local AI Advisor
Free · no sign-up · 48 models rated

Your own ChatGPT, running on your computer.

Local AI is private, free and works offline. Tell us what computer you have and we’ll show you which AI models will run smoothly on it, and exactly how to install them.

The check takes about a minute

Qwen 3.5 9B · on this laptop Offline

28 tok/s

Speed

0 bytes

Sent online

$$0

Per month

100%
free
Local AI in 30 seconds

Three pieces, all on your computer

You download a free app and an AI model. The app runs the model on your own machine. No account, no cloud.

Your computer
  1. 1. An appWhere you type and reade.g. LM Studio, Open WebUI, a coding agent
  2. 2. A runtimeThe engine that runs the modele.g. Ollama, llama.cpp, MLX
  3. 3. A modelThe AI itself: one big filee.g. Qwen, Gemma, Llama… 2–60 GB
Your question goes from the app to the runtime, which runs the model and streams the answer back. Nothing leaves your computer.
How it works

Three questions. Honest answers.

We do the math for your exact computer, so you don’t have to learn it first.

  1. 1Tell us your computerPick it from a list, or let us detect it for you. Mac, Windows PC or Linux.
  2. 2Say what you want to doChat, write, read long documents, code with an AI agent… each needs something different.
  3. 3Get models that feel goodRanked by how they’ll actually feel to use, with copy-paste install steps.
Why your computer matters

The same AI can type like a snail, or faster than you read.

Speed is measured in tokens per second (a token is about ¾ of a word). Tap the speeds to feel the difference. Your computer decides which one you get.

Learn what decides the speed
How can I save money on groceries?
Press play to watch the answer appear.

Readable About reading speed. Fine for chat.

0.0s / 7.8s
Why we’re different

“It fits” doesn’t mean “it runs well”.

Most calculators only check whether a model squeezes into memory. But your computer needs room to breathe too. We rate how a model will feel, from “Technically runs” to “Excellent”.

See why in 3 minutes
A laptop with 16 GB of memory18 / 16 GB
System
Your apps
Model
  • System · 4 GB
  • Your apps · 3 GB
  • Model · 9.5 GB
  • Chat memory · 1.5 GB

It “fits”… but your computer crawls

The model squeezes out your apps. The computer starts swapping to disk and everything, including the AI, slows to a crawl.

Popular computers

What runs on a computer like yours?

Our top picks for chatting and for coding on the machines people ask about most.

16 GB unified · 120 GB/s≈$999

MacBook Air M4 16 GB

Runs 11 of 48 models well enough for chat

Best for chat

MiniCPM5 2B · Q8_0

Excellent25–35 tok/s

Best for coding agents

MiniCPM5 2B · Q8_0

Acceptable20–30 tok/s

48 GB unified · 273 GB/s≈$2,799

MacBook Pro M4 Pro 48 GB

Runs 33 of 48 models well enough for chat

Best for chat

Qwen3.6 35B-A3B · 5-bit

Excellent60–80 tok/s

Best for coding agents

Qwen3.6 35B-A3B · 5-bit

Comfortable50–70 tok/s

128 GB unified · 546 GB/s≈$4,099

Mac Studio M4 Max 128 GB

Runs 41 of 48 models well enough for chat

Best for chat

gpt-oss-120b · MXFP4

Excellent85–110 tok/s

Best for coding agents

Qwen3.6 35B-A3B · 6-bit

Excellent80–110 tok/s

Swipe for more. Picks assume a chat app (Open WebUI) and a coding agent (OpenCode) on a medium-sized project.

Free course

New to local AI? Learn it in 10 short lessons.

Animated, jargon-free, about 37 minutes in total. Each lesson ends with a quick question so you know it clicked.

Start learning
  1. 1What is local AI?3 min
  2. 2Memory: where the model lives4 min
  3. 3How big is a model?4 min
  4. 4Quantization: shrinking models3 min
  5. 5Why “it fits” isn’t enough3 min
  6. 6Context: the model’s short-term memory4 min
Where to start

Pick the one that sounds like you

More tools for when you’re ready

Ratings come from a model of each computer’s memory and speed, calibrated against public benchmarks. How we calculate