OpenAI unveils DALL-E, a text-to-image generator

OpenAI introduced DALL-E on 5 January 2021, a 12-billion-parameter version of GPT-3 trained to generate images from text captions. It combined unrelated concepts, rendered text inside images and produced anthropomorphized objects and animals from written prompts.

Why it mattered The architecture and scaling behind GPT-3 turned out to generate images as well as text, a capability later image systems were built on.