Skip to content
EN
English 简体中文 soon 日本語 soon

CanIRun.ai

Check if your machine runs a model

Visit official site

What CanIRun.ai is

CanIRun.ai is a hardware compatibility checker for local AI models. It reads GPU, VRAM, RAM and CPU core information available to the browser and filters models into fit levels ranging from comfortable to too heavy, with filters for task type, provider and licence.

What you can do with it

  • See which models your machine should handle
  • Filter by chat, code, reasoning or vision
  • Compare providers
  • Check licence terms such as Apache 2.0 or MIT

Who it is for

  • Developers running models locally
  • People choosing hardware for inference
  • Teams comparing open models

What to watch out for

  • Browser estimates are not benchmarks: quantisation, runtime, driver versions, background load and context length all change real speed
  • Licence filtering is informative, not legal advice; read the actual licence before commercial use
  • Hardware details reported by a browser can be incomplete, especially for shared or virtual GPUs
  • Measure locally before committing to a deployment

Pros & cons

✓ What we like

  • Quick sanity check before downloading
  • Task and licence filters
  • No installation needed

! What to watch out for

  • Estimates, not benchmarks
  • Browser reporting can be incomplete
  • Licence details to verify

FAQ

What hardware does it detect?

GPU, VRAM, RAM and CPU core information available to the browser.

Are the results accurate?

They are estimates. Quantisation, runtime, drivers and context length affect actual performance.

Which providers are covered?

The site lists providers including Alibaba, DeepSeek, Google, Meta, Microsoft, Mistral, OpenAI and Qwen.

Last reviewed: 2026-09-17

More AI coding tools tools

View all →

How we review