# lucidrains/DALLE-pytorch

Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

Repository: https://github.com/lucidrains/DALLE-pytorch
Canonical: https://ross.abutalabs.com/products/dalle-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: artificial-intelligence, deep-learning, attention-mechanism, text-to-image, transformers, multi-modal
Last push: 2024-02-17T21:42:10+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": 2066, "days_push": 928, "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 5627, forks 642 (observed 2026-08-28T04:09:25.825993+00:00)

## What it is
A PyTorch implementation/replication of OpenAI's DALL-E, a text-to-image transformer, including a discrete VAE and optional CLIP for ranking generations. It is a research library for training and experimenting with DALL-E-style generative models.

## Use cases
- train a DALL-E text-to-image model on my own image-caption dataset
- generate images from text prompts in pytorch
- replicate the DALL-E paper for research
- experiment with transformer-based image generation
- train a small text-to-image model on a custom dataset
- use CLIP to rank generated images

## When to choose
- you want an open-source, hackable PyTorch implementation of DALL-E for research or experimentation
- you want to train a text-to-image transformer on your own dataset
- you want to study or modify the DALL-E architecture

## When to avoid
- you want state-of-the-art text-to-image generation out of the box - newer models like Stable Diffusion are far better
- you need a production-ready image generation service with pretrained large-scale checkpoints
- you don't have GPU resources for training

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, transformers, llm-training
- domain: artificial-intelligence, deep-learning, image-processing
- platform: python
- tags: text-to-image, dall-e, pytorch, generative-models, multi-modal, clip, vq-vae, natural-language-processing, gpu

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:25.825993+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 2026-08-29T17:55:21.507690+00:00, confidence not recorded.
  - readme: https://github.com/lucidrains/DALLE-pytorch (fetched 2026-08-28T04:09:25.825993+00:00, sha f0377e8d561a)
  - registry_pypi: https://pypi.org/pypi/dalle-pytorch/json (fetched 2026-08-29T08:49:57.116333+00:00, sha 2195d4bfa996)
- Data as of 2026-08-30T08:39:29.467469+00:00.