Pete Walsh

Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models