The DALL·E 2 — multimodal learning in image generation
The DALL·E 2 is the GPT-3 for images.
The DALL·E 2 is a text-conditional image generator based on the diffusion models and the inverted CLIP. Insert a text as an input. The DALL·E 2 will output an image matching the text.
The DALL·E 2 wait list was +100k people after its first week…