Versions compare:
v1.1 has extended LoRA & VAE loaders
v1.0 One LoRA, no VAE Loader, simple
Use ComfyUI manager for install missing nodes - https://github.com/ltdrdata/ComfyUI-Manager
Crisp and beautiful images with relatively short creation time, easy to use. SD and SDXL and Loras models are supported. The usage description is inside the workflow.
This workflow perfectly works with 1660 Super 6Gb VRAM
SD1.5 without lora, takes ~450-500 seconds with 200 steps with no upscale resolution (see workflow screenshot from ver 1.1)
Description
Compared to version 1.0, the changes are as follows:
Added LoRA Stack, you can enable up to 3 loras at the same time.
Added VAE loader, some models require it as a separate file, you can switch between the default VAE from the model or your own VAE.
Nodes with editable parameters are now highlighted in yellow. Updated workflow description.
FAQ
Comments (3)
Works great! Thank you.
Is there a way to select if you want only txt2img or img2img? I want to just upscale an already made image, for example. Is there a way to click (say,a button)? Like a Fast Bypasser hub (a node I saw users using?) Thanks!
doesn't work gives the following error:
VAE Input Switch
WAS_VAE_Input_Switch.vae_switch() got an unexpected keyword argument 'boolean_number'