PRIOR: Perceptual Reasoning and Interaction Research

High-impact open-source AI research

Featured Project

Molmo

Molmo is a family of open state-of-the-art multimodal AI models

Embodied AI

Project Image

SPOC

SPOC is a vision-language agent that follows language instructions to navigate, explore, and manipulate objects in real and virtual environments using only RGB input.

Learn More
Project Image

PoliFormer

PoliFormer is an RGB-based indoor navigation agent that supports long-term memory and reasoning for tasks like object navigation, tracking, and open-vocabulary exploration.

Learn More

Vision-Language

Project Image

Unified-IO 2

Unified-IO 2 is a single unified transformer that processes images, text, audio, and more, enabling seamless multimodal understanding and generation through a unified architecture.

Learn More
Project Image

One Diffusion

All-in-one diffusion model for text-to-image generation, conditional generation, image understanding, ID customzation and multi-view generation.

Learn More

Earth Systems

Project Image

Satlas

Satlas is an AI-powered platform that provides quarterly global updates on marine infrastructure, renewable energy sites, and tree cover using satellite imagery.

Learn More
Project Image

Galileo

Galileo is a family of generalist transformer-based satellite imagery models for remote sensing tasks like crop mapping, marine debree monitoring, and flood detection.

Learn More