Minirun

Runs models larger than memory on Mac and iPhone, streaming weights from SSD.

Mobile appfreeglobal

Minirun runs language models far larger than a device's available memory by streaming weights from SSD through a configurable memory budget, making it possible to run something like a 2.78-trillion-parameter Kimi K3 model on an iPhone 16 Pro, entirely free and local on Mac and iPhone. It targets developers and enthusiasts who want to run massive LLMs on Apple hardware without the RAM those models would normally require.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Minirun - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim Minirun to unlock →

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Minirun (or something else) actually fits.

Try a problem