Announcement_7

minDALL-E, a text-to-image generation model trained on 14M pairs, is now available at GitHub.