Very similar to my latent interposer, this small model can be used to upscale latents in a way that doesn't ruin the image. I mostly explain some of the issues with upscaling latents in this issue. Think of this as an ESRGAN for latents, except severely undertrained.

Currently, SDXL has some minimal hue shift issues. Because of course it does.

Installation

ComfyUI

To install it, simply clone this repo to your custom_nodes folder using the following command: git clone https://github.com/city96/SD-Latent-Upscaler custom_nodes/SD-Latent-Upscaler.

Alternatively, you can download the comfy_latent_upscaler.py file to your ComfyUI/custom_nodes folder as well. You may need to install hfhub using the command pip install huggingface-hub inside your venv.

If you need the model weights for something else, they are hosted on HF under the same Apache2 license as the rest of the repo.

Auto1111

Currently not supported but it should be possible to use it at the hires-fix part.

Usage/principle

Usage is fairly simple. You use it anywhere where you would upscale a latent. If you need a higher scale factor (e.g. x4), simply chain two of the upscalers.

As part of a workflow - notice how the second stage works despite the low denoise of 0.2. The image remains relatively unchanged.

Training

Interposer v1.0

This current version is still relatively undertrained, as with the interposer. Mostly a proof-of-concept but it seems good enough as a base.

Trained for 1M iterations on DIV2K + Flickr2K.

(Left is training loss, right is validation loss.)