DALL-E

tags
Transformers, GPT
paper
(Ramesh et al. 2021)

Architecture

It is a decoder architecture with a Variational autoencoders and a variant of GPT-3 to convert text to images.

Parameter count

12B

Bibliography

  1. . . "Zero-shot Text-to-image Generation". arXiv. DOI.

Links to this note

Last changed | authored by

Comments


← Back to Notes