# CompVis/stable-diffusion

A latent text-to-image diffusion model

Repository: https://github.com/CompVis/stable-diffusion
Canonical: https://ross.abutalabs.com/products/stable-diffusion
Homepage: https://ommer-lab.com/research/latent-diffusion-models/
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Last push: 2024-06-18T01:53:49+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1484, "days_push": 807, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 73347, forks 10576 (observed 2026-08-28T04:12:21.297544+00:00)

## What it is
The original reference implementation of Stable Diffusion, a latent text-to-image diffusion model trained on LAION-5B data with a CLIP text encoder and 860M UNet. It provides code for generating images from text prompts, plus inpainting and other image synthesis tasks, runnable on a GPU with ~10GB VRAM.

## Use cases
- generate images from text prompts
- run stable diffusion locally on my gpu
- inpaint or edit parts of an image with a diffusion model
- experiment with latent diffusion model code
- download and use stable diffusion v1 model weights
- do research on text-to-image diffusion models

## When to choose
- you want the original reference implementation and model weights for Stable Diffusion v1
- you need a relatively lightweight text-to-image model that fits in ~10GB VRAM
- you are doing academic research on latent diffusion models

## When to avoid
- you want a production-ready or actively maintained inference stack - use diffusers or newer Stable Diffusion releases instead
- you need the latest model versions like SDXL or SD3
- you have no GPU available
- your use case conflicts with the model's CreativeML OpenRAIL license restrictions

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, deep-learning
- domain: machine-learning, image-processing, artificial-intelligence
- platform: python, windows
- tags: stable-diffusion, text-to-image, diffusion-models, latent-diffusion, generative-ai, image-generation, research-code, gpu, linux, macos

## Member repositories
- CompVis/stable-diffusion (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:21.297544+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 2026-08-29T16:15:24.374668+00:00, confidence not recorded.
  - readme: https://github.com/CompVis/stable-diffusion (fetched 2026-08-28T04:12:21.297544+00:00, sha 8134717cf2c2)
  - homepage: https://ommer-lab.com/research/latent-diffusion-models/ (fetched 2026-08-28T17:46:30.132917+00:00, sha f9bd99e228c5)
- Data as of 2026-08-30T08:39:29.467469+00:00.