# openai/consistencydecoder

Consistency Distilled Diff VAE

Repository: https://github.com/openai/consistencydecoder
Canonical: https://ross.abutalabs.com/products/consistencydecoder
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2023-11-07T11:21:38+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 73
- inputs: {"age_days": 1035, "days_push": 1030, "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 2212, forks 81 (observed 2026-08-28T04:06:26.895793+00:00)

## What it is
A Python library from OpenAI implementing a consistency-distilled diffusion VAE decoder that improves image decoding quality for Stable Diffusion VAE latents. It integrates with Hugging Face diffusers and PyTorch to replace the standard GAN decoder with higher-fidelity decoding.

## Use cases
- decode stable diffusion latents with better image quality
- replace the GAN decoder in a stable diffusion VAE pipeline
- improve fidelity of generated images from stable diffusion
- run consistency models for image decoding
- integrate an improved VAE decoder into diffusers workflows

## When to choose
- you use Stable Diffusion and want higher-quality decoded images from latents
- your pipeline already uses PyTorch and Hugging Face diffusers
- you have a CUDA GPU with memory for the ~2.5 GB model

## When to avoid
- you need CPU-only inference or very low memory environments
- you use a different image generation stack without stable diffusion VAE latents
- you need actively maintained software with recent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: image-processing, machine-learning, artificial-intelligence
- platform: python
- tags: stable-diffusion, vae-decoder, consistency-models, diffusers, image-generation, gpu

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:26.895793+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 2026-08-30T02:46:02.883740+00:00, confidence not recorded.
  - readme: https://github.com/openai/consistencydecoder (fetched 2026-08-28T04:06:26.895793+00:00, sha c758298d862d)
- Data as of 2026-08-30T08:39:29.467469+00:00.