Short on time, description is short for now. I will update it when I get the chance!
Yet another model born from a series of tests I've been performing to evaluate a dataset creation toolkit I've been working on.
This model is trained only on high quality anime illustrations, screencaps, digital art.
Instead of strictly using booru tags, the dataset was captioned with an expanded tagging database connected to an MLLM (more info on this soon).
Buy me a coffee ❤
https://ko-fi.com/ndimensional
All donations will be used to fund the creation of new Stable Diffusion fine-tunes and open-source AI tools.
Check out my other models
SDXL Checkpoints: https://civarchive.com/collections/966964
SDXL LoRAs: https://civarchive.com/collections/966969
40K Series: https://civarchive.com/collections/956187
SD1.5 Checkpoints: https://civarchive.com/collections/966974
SD1.5 LoRAs: https://civarchive.com/collections/966972
Description
Initial Release
FAQ
Comments (4)
This look very good but i can't reproduce any of your images. They're similar but different. Do you use any extensions or else? I use A1111.
CFG rescaler could be the cause, though it's not enabled. Could also be an issue with me using DPM++ samplers. I'll take a look at it when I get off work 👍
Well done on making such a great model! Very competently trained, you have done a great job with the model so far. It appears to understand prompts marginally better than other models do too. Great job!
I was slightly concerned by the Trump Reference in the image samples tho... 👀
\(female\), \(style\)
What does it mean?