Please Donate Buzz for FLUX Lora Training !

FLUX v1.0 :

Trained on FLUX.1 [dev] with 216 photos of Elizabeth Lail. 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:elilail_flux_1_standard_v2-000019:1>

v2.0 :

Trained on Juggernaut X with 216 photos of Elizabeth Lail with an improved Lora training method for better likeness and flexibility. Tested on Juggernaut X, Juggernaut v7, RealismEngine 2, RealVisXL3, AlbedoBaseV2! Use with keyword : "elxlail" . 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}, elxlail, <lora:elillail_juggerX_xl_2_st_merger_20_98_035_065-elxlail:1>

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

v1.0

Lora trained on 210 images of American actress Elizabeth Lail (Vanessa in FNAF movie and Beck in You) and captioned with GPT4-V and trained on SDXL 1.0 Base. Recommended strength is around 1.1. But you should experiment yourself with your preffered model. Tested on Juggernaut XL 3, Juggernaut XL 7, RealVision XL 2, DynaVision XL, SDXL Base.