wowsuch ai
Live

YidharModel

psd-layer-vae h3

psd-layer-vae h3 is an RGBA autoencoder for editable, layered PSD images, encoding each layer into its own latent representation and decoding layers jointly. It supports HDR reconstruction and is described as the default variant for editing and layer-by-layer fidelity.

What it does

Use it to encode and reconstruct layered PSD artwork while preserving layer transparency, fine strokes, and HDR information.

Availability
Repository is public; access to files requires agreeing to conditions and sharing Hugging Face information.
Pricing model
Not established
Deployment
Inference code and weights are provided; example uses CUDA.
Privacy
Not established
Licence
Apache License 2.0 for weights and inference code.
Capability
image generation
Modality
image
Access
open weightsSDK
Deployment
on-premises
Pricing
unknown
Target user
developercreatorresearcher

Evidence

5 claims
  • Each layer is encoded into a latent with 1/8 resolution and 32 channels. [1]
  • The decoder uses cross-layer attention and outputs straight RGBA. [1]
  • An HDR channel can reconstruct linear light above SDR white. [1]
  • The repository is publicly accessible, but files require agreeing to conditions and sharing Hugging Face information. [1]
    This repository is publicly accessible, but you have to accept the conditions to access its files and content .
    Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The h3 variant is designated for editing and layer-by-layer reconstruction. [1]

Sources

1
1
Hugging FaceOwner-controlled
https://huggingface.co/Yidhar/psd-layer-vae Retrieved 2026-11-02 01:04 UTC
Verified

Versions

1
VersionEffectiveChangeChanged fieldsState
12026-11-02 01:04 UTCpublicationNoneCurrent

Open in the catalog

Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.