Molmo2 ER is a multimodal model from the Allen Institute for AI that processes both images and text, producing text-based responses. It carries the open-weight, Apache-licensed ethos of AllenAI's research lineage — transparent and accessible for inspection and modification. Specific capability details beyond its multimodal input handling are limited in available documentation.