Hacker News with Generative AI: Multimodal Models

Mistral releases Pixtral 12B, its first multimodal model (techcrunch.com)
French AI startup Mistral has released its first model that can process images as well as text.
Generative AI, Artificial Intelligence, France, Multimodal Models, Language Models
163 points by jerbear4328 83 days ago | 40 comments
Transfusion: Predict the next token and diffuse images with one multimodal model (arxiv.org)
We introduce Transfusion, a recipe for training a multi-modal model over discrete and continuous data.
Generative AI, Machine Learning, Multimodal Models, Computer Vision
122 points by fzliu 85 days ago | 10 comments