RDBT | Krea2
Improve aesthetic, details. More natural, less slop-ish. Supports both photograph and anime art.
Dataset contains ~4k images. All handpicked. Zero shiny plastic glossy AI image. All captions are NL from LLM.
Update log:
v1.0: All-in, full dataset.
v0.2: New method. Refined dataset and captions. x2 more images. Also a professional photograph enhancer. (If you start to rent a h200 node, why not dump more data, as it's already as expensive as f*** ...)
v0.1: 1k images, 12 epochs, all default settings, testing.
Sharing merges using this model is not allowed. This is a free model and it will always free. This "restriction" won't affect anyone, it's only aimed at those who steal others' models to sell.
FYI: all my models have special trigger words and latent watermark.
Description
FAQ
Comments (6)
Interesting, so this is a complement to the Turbo/Raw+Turbo workflows to enhance anime aesthetic and concepts?
He thought it would be fine to not explain whats the purpose of this lora whatsoever
Please can you do a realism enhancer checkpoint or LORA like your great chroma one maybe integrating the Krea2FilterBypass and Realism Engine Ideogram 4 + Krea 2 Lora too.
No. I don't do realism. Besides, there are many good LoRAs/finetunes for realism for those models.
heyy, what would you say about krea2? Does it have a potential to become better than anima or not
Depends.
If you can train LoRA that you need, for characters and styles, and want more creative concepts. krea2 is a beast, not even close. I would say you should definitely switch to krea2.
But for most people, anima is more handy for out-of-the-box characters and styles, and the quality is good enough. and small.