PsyFi XL is a merge of 17 checkpoints & 26 loras, with a specific focus on psychedelic, surreal, and/or sci-fi generations. That being said, its flexibility extends beyond these focal areas.
The current version is a starting point for additional merging and training. There are many rooms for improvement I have already identified, but they are far outnumbered by the rooms I have not yet entered. I appreciate any/all feedback on avenues to improve upon!
Below are links to all of the models implemented in this merge. Eternal appreciation and thanks to all of these creators!
Checkpoints Used:
Loras Used:
Description
FAQ
Comments (7)
Love this style.
My new favorite model !
I don't know what it is, but I get a lot of watermarks/signatures in the lower right unless I keep the refiner set to a very high percentage
For some reason, I keep getting this error when I try to load the model:
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 14.00 MiB. Any ideas how I could solve this? Mycard is GF RTX2060, 6GB
What application do you use? I have better luck when using ComfyUI for SDXL models on my RTX3060, 12GB
@Cinemachina I'm using SD webui
This one is incredible! I've been experimenting a lot for the past week with models to upscale my psychedelic images i get out of Bing inside StableDiffusion.
This is the only model so far which can really add to the magic. I love it!
Wow thank you, this is amazing to hear! Glad it is helping out!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.