Thanks to everyone who supported and donated buzz, v0.2 was trained on Civitaiv02 brings: fixed some artefacts, better hands &amp; feet, more eye details, better colors<hr />This is my first model, well the first iteration and I wasn't planning on publishing it but it came out pretty good. I am quite particular about certain color styles so this is why I decided to train my own model.This was trained using [fluxgym](<a target="_blank" rel="ugc" href="https://github.com/cocktailpeanut/fluxgym">https://github.com/cocktailpeanut/fluxgym</a>) with a 22 images dataset and 10 Epochs.base_model: black-forest-labs/FLUX.1-devprompt: in the style of dmai (works well with short and long prompts)weight: 0.4 - 1.5 after it, it starts being quite too much but you are free to try different settingsCFG: 1steps: 20 - 37You can buy me a coffee ˗ˏˋ☕ˎˊ˗ <a target="_blank" rel="ugc" href="https://ko-fi.com/dan81mai">here</a> ❤️

Files with SHA256: 3bff1956a15c9d84096e38a6f40688c0ed545e2d59978f3aa8d14cf81a26ebb4