Our demos for anyone wanting to experience our latest AI research breakthroughs firsthand.

Featured demo

SAM can “cut out” any object, in any image, with a single click

The new Segment Anything Model (SAM) is a promptable segmentation system with zero-shot generalization to unfamiliar objects and images, that has no need for additional training. An accompanying dataset, SA-1B, is also being released as part of the launch.


Computer vision


A family of models producing universal features suitable for image-level visual tasks as well as pixel-level visual tasks.

Computer vision


Computer vision powered by self-supervised learning to build highly efficient systems that perform well with less labeled data.

Computer vision

Animated Drawings

A first-of-its-kind method for using AI to animate amateur drawings


ESM Metagenomic Atlas

An open atlas of 772 million predicted metagenomic protein structures

Machine translation

No Language Left Behind

Stories Told Through Translation: books from around the world translated into hundreds of languages with the NLLB-200 model.