# LTH14/fractalgen

PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437

Repository: https://github.com/LTH14/fractalgen
Canonical: https://ross.abutalabs.com/products/fractalgen
Language: Python
License: MIT
License Family: permissive
Last push: 2025-02-25T12:12:51+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 8, release rhythm 35, longevity 40
- inputs: {"age_days": 561, "days_push": 554, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1244, forks 67 (observed 2026-08-28T04:04:06.771852+00:00)

## What it is
A PyTorch implementation of Fractal Generative Models (FractalGen), enabling pixel-by-pixel high-resolution image generation. It includes pre-trained models on ImageNet 64x64/256x256, a Colab demo, and DDP training/evaluation scripts.

## Use cases
- generate high-resolution images pixel by pixel
- run pretrained fractal generative models in a colab notebook
- train an autoregressive image generation model on imagenet
- reproduce results from the fractal generative models paper
- experiment with novel generative model architectures in pytorch

## When to choose
- you want to reproduce or extend the FractalGen paper
- you need pixel-by-pixel high-resolution image generation with pretrained checkpoints
- you want a simple PyTorch research codebase for generative models

## When to avoid
- you need production-ready image generation APIs
- you lack GPU resources for large 400M+ parameter models
- you need text-to-image or diffusion-based generation

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, llm-training
- domain: deep-learning, machine-learning, image-processing, artificial-intelligence
- platform: python
- tags: pytorch, generative-models, image-generation, research-code, pretrained-models, autoregressive, fractal, gpu

## Member repositories
- LTH14/fractalgen (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.771852+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 2026-08-30T05:08:13.252070+00:00, confidence not recorded.
  - readme: https://github.com/LTH14/fractalgen (fetched 2026-08-28T04:04:06.771852+00:00, sha c382fa1d338e)
- Data as of 2026-08-30T08:39:29.467469+00:00.