# lucidrains/DALLE2-pytorch

Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network,  in Pytorch

Repository: https://github.com/lucidrains/DALLE2-pytorch
Canonical: https://ross.abutalabs.com/products/dalle2-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: artificial-intelligence, deep-learning, text-to-image
Last push: 2024-05-11T19:18:10+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 1609, "days_push": 844, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 11306, forks 1070 (observed 2026-08-28T04:10:46.692593+00:00)

## What it is
A PyTorch implementation of OpenAI's DALL-E 2 text-to-image synthesis model, focusing on the diffusion prior network that predicts image embeddings from CLIP text embeddings. It provides training scripts and model components for researchers replicating the architecture.

## Use cases
- generate images from text descriptions
- train a diffusion prior for CLIP embeddings
- replicate DALL-E 2 architecture in PyTorch
- research text-to-image synthesis models
- build a CLIP-to-StyleGAN text-to-image pipeline
- experiment with diffusion-based image generation

## When to choose
- you want to study or reproduce the DALL-E 2 architecture in PyTorch
- you need a diffusion prior network for CLIP-based image generation research
- you want a readable reference implementation of the paper

## When to avoid
- you need state-of-the-art text-to-image generation - the author recommends Imagen instead
- you want a production-ready model with official OpenAI weights
- you need a simple plug-and-play image generator without training your own models

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, stable-diffusion, image-processing
- domain: artificial-intelligence, deep-learning, machine-learning, image-processing
- platform: python
- tags: text-to-image, dalle2, diffusion-models, pytorch, clip, generative-ai, research-implementation, gpu

## Member repositories
- lucidrains/DALLE2-pytorch (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:46.692593+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:15:58.410267+00:00, confidence not recorded.
  - readme: https://github.com/lucidrains/DALLE2-pytorch (fetched 2026-08-28T04:10:46.692593+00:00, sha c137711953a5)
- Data as of 2026-08-30T08:39:29.467469+00:00.