You can click on the link below to try it out directly. If the effect is good, you can deploy it locally
https://www.runninghub.ai/post/1955386783511056385/?inviteCode=1cqzbf7a
Fan benefits,register to get 1000 points,daily login 100 points,play 4090!Experience the super power of 48G.
I used a lot of animation and reality fused images for training, and personally I really like this style. I hope everyone likes it
Suggest adding the following prompt words:
The perfect fusion of anime characters and real-life scenes.
Description
FAQ
Comments (4)
The base model knows both anime and realistic styles, obviously. So the real question is what should be the ideal training for this LoRA? Surely it should be on the prompt side, where the prompt can tag either 'anime' or 'real' for various scene elements. I don't understand your prompt concept at all.
Besides, surely this is a task for inpainting via masking?
This is a SUPER interesting new style. I love it!
This is my favorite all time lora thus far - of ANY model I've tried to date.
It is highly underrated - I'm quite surprised it hasn't garnered more attention!!
Thanks!! :)
Wonderful model! Any plans for a Z-Image release?