# lucidrains/denoising-diffusion-pytorch

Implementation of Denoising Diffusion Probabilistic Model in Pytorch

Repository: https://github.com/lucidrains/denoising-diffusion-pytorch
Canonical: https://ross.abutalabs.com/products/denoising-diffusion-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: artificial-intelligence, deep-learning, generative-model, score-matching
Last push: 2026-08-02T20:11:09+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 95, release rhythm 70, longevity 100
- inputs: {"age_days": 2199, "days_push": 31, "days_rel": 203, "gap_med": 1.5, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10679, forks 1281 (observed 2026-08-28T04:10:43.737338+00:00)

## What it is
A PyTorch implementation of Denoising Diffusion Probabilistic Models (DDPM) for generative image modeling, published on PyPI as denoising_diffusion_pytorch. It provides Unet and GaussianDiffusion components plus a Trainer class for training diffusion models on image folders.

## Use cases
- train a diffusion model to generate images
- implement DDPM in pytorch
- learn how denoising diffusion probabilistic models work
- generate images from a folder of training data
- experiment with DDIM sampling and diffusion timesteps
- build a generative image model without GANs

## When to choose
- you want a simple, readable PyTorch codebase for training diffusion models on images
- you need a pip-installable DDPM implementation with a built-in Trainer
- you are researching or prototyping diffusion-based generative models

## When to avoid
- you need production-grade, feature-complete diffusion tooling like text-to-image pipelines or schedulers
- you work outside PyTorch or need non-image modalities
- you require enterprise support or long-term stability guarantees

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing
- domain: deep-learning, machine-learning, artificial-intelligence, image-processing
- platform: python
- tags: diffusion-models, generative-models, pytorch, ddpm, score-matching, image-generation, research-code, gpu

## Member repositories
- lucidrains/denoising-diffusion-pytorch (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:43.737338+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 2026-08-29T17:18:11.607692+00:00, confidence not recorded.
  - readme: https://github.com/lucidrains/denoising-diffusion-pytorch (fetched 2026-08-28T04:10:43.737338+00:00, sha cbdca235b2f0)
  - registry_pypi: https://pypi.org/pypi/denoising-diffusion-pytorch/json (fetched 2026-08-29T08:17:12.928466+00:00, sha 129187b49d71)
- Data as of 2026-08-30T08:39:29.467469+00:00.