RDBT | Krea2

RDBT | Krea2

Improve aesthetic, details. More natural, less slop-ish. Supports both photograph and anime art.

Dataset contains ~4k images. Professional photograph + digital anime art. All handpicked. Zero shiny plastic glossy AI slop image. All captions are NL from LLM.

Strength 1 is recommended. Dataset is big enough and model is not overfitted. Normally it won't drastically change image.

Update log:

v2.3 (kroma v0.3):

Based on kroma v0.3. Testing. This version is beardly trained, but it works.

Not a big update. The version jumped to v2.3 because I just want to align the version number to my dataset and my other RDBT models.

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.