Moondream2

Moondream2 is a compact open-source vision-language model designed for efficient on-device and edge inference. It accepts image and text inputs to perform visual question answering, image captioning, object detection, and region pointing. Source: supabase.