The LoRA was trained on Flux-Dev with one image as an experiment and might behave a little bit erratic.Please check out the prompts of the showcase images and follow the prompting patternIt might not work with other Flux Versions. If it works, expect it to behave differently then with Flux-Dev.The showcase images are made with Flux-Dev.For Flux Dev I recommend the following setting - Lora strength 0.8-1.0, highres fix with denoising 0.35-0.4