LocalModelMatchnew models. same place.
local AI, with a little chemistry.

Your computer
has a type.

Good chemistry. Real memory limits.

A meet-cute for your computer.

What computer are we matching?

Choose your target computer, even if you’re visiting on a phone. No account or download.

New releases. Fresh fits.

Automatic checks for new and trending models on every visit, and every 15 minutes while open. Major releases enter through the same live feeds as new fine-tunes and quants.

How updates work

Available MLX and CUDA variants appear as their public metadata is discovered. New releases may take time to gain compatible variants or benchmark scores. Benchmark sources refresh every 6 hours; unavailable sources keep their last dated results.

You’re exploring an example setup

Confirm your computer’s memory before relying on these results.

49 estimated fits · 93 matching variantsMLX catalog

Memory decides what fits. Intelligence is a published test score. Speed is an estimated output rate. Open any model for Hugging Face links, install instructions, and the evidence.

MODEL / VARIANTINTELLIGENCEEST. OUTPUTMEMORYMATCH
Saved catalog. Checking for updates…Metadata checked Sep 6, 2026

25 sourced benchmark results · Benchmark sources refresh every 6 hours on use.