# lllyasviel/LayerDiffuse

Transparent Image Layer Diffusion using Latent Transparency

Repository: https://github.com/lllyasviel/LayerDiffuse
Canonical: https://ross.abutalabs.com/products/layerdiffuse
License: Apache-2.0
License Family: permissive
Last push: 2024-06-16T08:05:00+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 65
- inputs: {"age_days": 918, "days_push": 808, "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 2221, forks 36 (observed 2026-08-28T04:06:27.642092+00:00)

## What it is
LayerDiffuse is a research project that generates transparent images and image layers using diffusion models with latent transparency. It provides entry points for Stable Diffusion WebUI (Forge), a Diffusers CLI, and plans for Gradio/HuggingFace integrations.

## Use cases
- generate transparent png images with stable diffusion
- create layered image assets for design
- remove backgrounds by generating transparent layers
- generate foreground and background layers separately
- integrate transparency into diffusion workflows

## When to choose
- you need AI-generated images with true transparency
- you want to generate layered compositions with diffusion models
- you use Stable Diffusion WebUI Forge or Diffusers

## When to avoid
- you need production-grade stable tooling
- you require non-Stable-Diffusion model support
- you need training code or dataset access before release

## Facets
- artifact type: library
- maturity: experimental
- function: image-processing, machine-learning, deep-learning, stable-diffusion
- domain: artificial-intelligence, image-processing, graphics
- platform: python, cross-platform
- tags: diffusion, transparent-images, latent-transparency, layer-generation, stable-diffusion, gpu

## Member repositories
- lllyasviel/LayerDiffuse (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:27.642092+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 2026-08-30T02:45:47.532627+00:00, confidence not recorded.
  - readme: https://github.com/lllyasviel/LayerDiffuse (fetched 2026-08-28T04:06:27.642092+00:00, sha ee4ce5b3396c)
- Data as of 2026-08-30T08:39:29.467469+00:00.