Please Donate Buzz for FLUX Lora Training !

FLUX v1.0 :

Trained on FLUX.1 [dev] with 75 photos of Scarlett Johansson 2004-2009 captioned with GPT-4 VIsion. 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},<lora:Scarlett_Johansson_FLUX_v1-000058:1>

v2.0 :

Trained on Juggernaut X with 230 photos of Scarlett Johansson 2004-2009 with improved likeness due to consistency and an improved Lora model. Tested on Juggernaut X, Juggernaut v7, RealismEngine 2, RealVisXL3, AlbedoBaseV2! Use with keyword : "scxjhnssn" . Use around strength 1.0. CFG 5.0-7.0. Clipskip 1. Can be used for example as follows:

Positive : {Artstyle}, {Character and scene description}, scxjhnssn, <lora:scjohansson_juggerX_xl_1_st_woc-scxjhnssn-000075:1.0>

Negative : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child

v1.0:

Lora trained on 300 (mainly portrait form) images of Scarlett Johansson with the SDXL 1.0 base version. Good at Portrait imagery. The recommended setting for ComfyUi is 0.9-1.1 Lora strength.