What is this?
This is the official PyTorch package for the discrete VAE (dVAE) used in OpenAI's DALL·E image generation model. It only contains the image encoder/decoder component, not the transformer that actually generates images from text.
Is this the full DALL·E model or just a piece of it?
It's just the discrete VAE component that compresses images into tokens and reconstructs them. The text-to-image transformer that generates images from prompts is explicitly not included in this release.
Do I need Python and PyTorch set up first?
Yes, it's a Python package built on PyTorch, installed via pip.
Can I generate images from text with this repo alone?
No, since the transformer that turns text into images is not released here, you only get the encode/decode (image-to-tokens and tokens-to-image) functionality.
Is this repository actively maintained?
No, the repository is archived, so it's a frozen historical release rather than an actively developed project.
Where can I see how to use the model?
The repo includes a usage.ipynb notebook demonstrating how to load and run the discrete VAE.
What are similar projects?
DALLE2-pytorch, shap-e, dalle-mini, latent-diffusion, nanoGPT, generative-models, PyTorch-GAN, stable-diffusion
Which topics is it in?
vision, text-to-image, gpt, neural-network, transformer, image-generation, generative-ai, deprecated, computer-vision, pytorch, openai, deep-learning