<h3 id="flux-v1.0-:-iu4nvwxvd">FLUX v1.0 :</h3>Trained on FLUX.1 [dev] with 110 photos of Kaia Scodelario. 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:Kaya_Scodelario_FLUX_v1-000055:1&gt;<h3 id="sdxl-v2.0:-c1spcd8en">SDXL v2.0: </h3>Lora trained on 110 images of Kaia Scodelario, captioned in natural language with GPT-4 Vision and trained on the SDXL 1.0 Base model with an improved Lora model. The strength should be between 0.8-1.2.<h3 id="sdxl-v1.0:-efvwviz77">SDXL v1.0: </h3>Lora trained on 60 images of actress Kaya Scodelario trained and tested on the SDXL 1.0 base model.

Kaya Scodelario (FLUX + SDXL)

Lora trained on 110 images of Kaia Scodelario, captioned in natural language with GPT-4 Vision and trained on the SDXL 1.0 Base model with an improved Lora model. The strength should be between 0.8-1.2.

Files with SHA256: 660d6965821633d0f5fbc9e966c67c329eeffd06cba4cd5d0b57678ebfb92fad