<h2 id="please-donate-buzz-for-flux-lora-training-!-9gfgdro2t">Please Donate Buzz for FLUX Lora Training !</h2><h3 id="flux-v1.0-(2023-2024):-kgoh0j0pa">FLUX v1.0 (2023-2024):</h3>Trained on FLUX.1 [dev] with 110 photos of Dua Lipa in 2023-2024. Tested on FLUX 1.D (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.8-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion}, &lt;lora:Dua_Lipa_new_FLUX_v2-000055:1&gt;<h3 id="sdxl-v1.0-0les6ahim">SDXL v1.0</h3>Lora trained on 130 images of British-Albanian Singer Dua Lipa. Captioned with GPT-4 Vision and trained with a slightly larger Lora model than usual. Lora strength should be between 0.7-1.2.

Files with SHA256: 7cfd1d11c8f5c5d59a4a011376cd3469e309252f7a838b46b08991554f8840da