Multimodal models

Molmo is a new series of multimodal vision-language models (VLMs) created by researchers at the Allen Institute for AI and the University of Washington. The Molmo family outperforms many state-of-the-art…