# openai/glow

Code for reproducing results in "Glow: Generative Flow with Invertible 1x1 Convolutions"

Repository: https://github.com/openai/glow
Canonical: https://ross.abutalabs.com/products/openai-glow
Homepage: https://arxiv.org/abs/1807.03039
Language: Python
License: MIT
License Family: permissive
Topics: paper
Archived: true
Last push: 2024-07-23T14:56:11+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2997, "days_push": 771, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3185, forks 524 (observed 2026-08-28T04:07:48.015598+00:00)

## What it is
Official research code from OpenAI for reproducing the 'Glow: Generative Flow with Invertible 1x1 Convolutions' paper, a flow-based generative image model built in TensorFlow 1.x. It includes training scripts, pretrained CelebA-HQ models, and an interactive demo for image synthesis and manipulation.

## Use cases
- reproduce results from the Glow generative flow paper
- train a normalizing flow model on CIFAR-10 or ImageNet
- generate and manipulate faces with a pretrained CelebA-HQ model
- run the interactive image manipulation demo
- benchmark log-likelihood on standard image datasets
- study invertible 1x1 convolution implementations

## When to choose
- you need to reproduce or extend the Glow paper's results
- you want a reference implementation of normalizing flows with invertible 1x1 convolutions
- you want to experiment with the pretrained CelebA-HQ face manipulation demo
- you are researching flow-based generative models

## When to avoid
- you need a maintained library or modern TensorFlow 2.x support
- you want a production-ready generative image model
- you lack multi-GPU Horovod/MPI infrastructure for large-scale training
- you need state-of-the-art image generation quality

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, image-processing
- domain: machine-learning, deep-learning, artificial-intelligence, image-processing
- platform: python
- tags: generative-models, normalizing-flows, tensorflow, research-code, paper-reproduction, archived, gpu

## Member repositories
- openai/glow (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:48.015598+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 2026-08-29T18:45:00.884758+00:00, confidence not recorded.
  - readme: https://github.com/openai/glow (fetched 2026-08-28T04:07:48.015598+00:00, sha d27f059573e1)
  - homepage: https://arxiv.org/abs/1807.03039 (fetched 2026-08-29T09:38:59.724838+00:00, sha 738d8707048b)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T09:38:59.727556+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T09:38:59.731284+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T09:38:59.733065+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T09:38:59.729600+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.