# openai/consistency_models

Official repo for consistency models.

Repository: https://github.com/openai/consistency_models
Canonical: https://ross.abutalabs.com/products/consistency_models
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2024-03-22T06:56:31+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 91
- inputs: {"age_days": 1284, "days_push": 894, "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 6486, forks 432 (observed 2026-08-28T04:09:43.762255+00:00)

## What it is
Official PyTorch implementation of Consistency Models, a generative image model family from OpenAI supporting consistency distillation, consistency training, and one- or few-step sampling. It includes pre-trained checkpoints for ImageNet-64 and LSUN 256 experiments.

## Use cases
- generate images in one or few steps with consistency models
- run consistency distillation on diffusion models
- train consistency models from scratch
- download and sample from pretrained ImageNet and LSUN checkpoints
- reproduce research experiments on generative image models

## When to choose
- you want fast few-step image generation without iterative diffusion sampling
- you need the official reference implementation for research on consistency models
- you want pretrained EDM/CD/CT checkpoints for ImageNet-64 or LSUN

## When to avoid
- you need a production-ready image generation service or API
- you want to generate high-resolution or text-to-image outputs beyond the released 64/256px models
- you need actively maintained code with recent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, data-generation
- domain: deep-learning, machine-learning, image-processing, artificial-intelligence
- platform: python
- tags: diffusion-models, generative-models, pytorch, consistency-distillation, consistency-training, image-generation, research-code, pretrained-models, gpu, docker, linux

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:43.762255+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 2026-08-29T17:44:32.337514+00:00, confidence not recorded.
  - readme: https://github.com/openai/consistency_models (fetched 2026-08-28T04:09:43.762255+00:00, sha 5cc40c4b0c95)
- Data as of 2026-08-30T08:39:29.467469+00:00.