Note: This upscaler is not mine
Credit to Helaman
Originally uploaded at: https://openmodeldb.info/models/4x-Nomos8kHAT-L-otf
About version 2.0
Everything is the same is with the first upload, but converted to .safetensors. I had issues getting Forge and Automatic1111 to load the .safetensors version of the upscaler but it works like a charm in ComfyUI.
General info
Hybrid Attention Transformer (HAT) combines channel attention and self-attention schemes and makes use of their complementary advantages. To enhance the interaction between neighboring window features, an overlapping cross-attention module is employed in HAT. Read more
Where does it go?
To use this (and other HAT upscalers) with Automatic1111 and Forge follow these steps.
Create a folder in in \webui\models\ and name it HAT
Download the file either here or from the source
Place the file in \webui\models\HAT\
Restart your webui
Note: If you have issues getting the model to work, change the file name from .pt to .pth
Description
The model and everything is the same as previous upload, but converted to .safetensors
Note: I had issues getting Forge/Automatic1111 to load the upscaler, but it works like a charm in ComfyUI
FAQ
Comments (7)
what's the difference with "4xNomos8kSCHAT-L"
I have no idea. You would have to ask the creator of the models.
It maybe a bit slow, but it is an excellent upscaler. It retains color, texture of images really well.
It's one of the best ones I have tried :)
good one
Can I use with ComfyUI?
Yes, just put it in the upscaler folder and load with any "Load upscaler" node.