JollyJack_LoRA_Alpha

V0.2 updated

Single character dataset with masked watermark.

Trained to 2250 steps

This is my very first attempt at making a LoRA for Flux Dev.

Trained using https://github.com/kijai/ComfyUI-FluxTrainer

AdamW - Cosine with Restarts

64 Dim

64 Network

Learning rate 3e-4

1600 steps (200 epoch) on 8 random images of jollyjack's art.

Seems to work best at 0.7 strength.