aramintak / linnea-flux-beta

An original character LoRA

  • Public
  • 182 runs
  • Fine-tune

Input

Output

Run time and cost

This model runs on Nvidia H100 GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

Model description

Meet Linnea!

This is my personal OC, Linnea, who is a big focus in my own art and illustration. The dataset is composed entirely of hand drawn illustrations and I'll be using her for some future projects. This is an early model, and I expect I'll work out some improved parameters in the near future so it should continue to get better.

I wanted to share her and will be writing some blogs/making videos in the near future on how to train a character like this. I also have no issue with people using her in their own experiments, particularly to test out mixing characters with styles for Flux Dev.

That being said, she is not For Commercial Use! This is only for fun and research. :) I would prefer not seeing her become someone's mascot. If you would like to collaborate on a project that features this character, you'll need to contact me and I might not agree to it anyway. :)

Note: The hair color still sometimes varies, it is easiest to fix if you use the prompt "teal hair".

Dataset example images: IMG_0093.PNG

IMG_0160.PNG

IMG_0208.PNG

Special Thanks!

I trained this model while I was testing out the Replicate training pipeline, big thanks to the Replicate team. :)

out-0 (3).png

Trigger words

You should use linnea teal hair to trigger the image generation.