A Deepdive into Aya Vision: Advancing the Frontier of Multilingual Multimodality

In three linesCohere introduces Aya Vision, a multimodal multilingual model processing images and text across 119 languages. The model combines vision and language understanding for image captioning, visual question answering, and document analysis tasks in low-resource languages.

Summary generated by Claude — human-verified