Model by nvidia

llama-nemotron-embed-vl-1b-v2

Multimodal question-answer retrieval representing user queries as text and documents as images.