Please Donate Buzz for FLUX Lora Training !
FLUX v1.0 :
Trained on FLUX dev with 80 photos of Jenna Coleman with detailed GPT-4 captions. Tested on FLUX 1.dev (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.8-1.0. 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:Jenna_Coleman_FLUX_V1-000061:1>
SDXL v4.0 :
Trained on Juggernaut X with 210 photos of Jenna Coleman with improved likeness due to consistency and an improved Lora model. Tested on Juggernaut X, Juggernaut v7, RealismEngine 2, RealVisXL3, AlbedoBaseV2! Use with keyword : "jenxcolemn" . Use around strength 1.0-1.1. CFG 5.0-7.0. Clipskip 1. Can be used for example as follows:
Positive : {Artstyle}, {Character and scene description}, jenxcolemn, <lora:jencoleman_juggerX_xl_1_wocap_merger_23_145_merger_106_045_055-jenxcolemn:1>
Negative : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child
SDXL v3.0 160mb :
Trained on 130 photos of Jenna Coleman on SDXL 1.0 base with better Lora presets. Less than half the size of the previous version. Use with keyword : "jenxcoleman"! Lora strength with prompt should be around 1.0. Following prompt structure on Juggernaut v7, realismEngine 2.0 and Juggernautv9 lightning have been tested and work well:
positive : {art style}, {character descrioption and action},jenxcoleman, <lora:jencoleman_32_small_xl_7_standard_wo_cap-merger-29_52_04_06-jenxcoleman:1.0>
negative : photoshop, rendered, (multiple people), blurry, child
SDXL v2.0 :
Trained on 200 images with an improved Lora model to ensure better likeness and flexibility. Keep lora strength near 1.0-1.1.
SDXL v1.0 :
Lora of British actress Jenna Louise Coleman, trained on 50 images of her on the Base SDXL model. The tested strength on ComfyUi is 0.9-1.2.
