Toy Box Flux LoRA

Toy Box Flux was trained entirely on AI-generated images. The concept was to blend the weights of an existing 3D LoRA by SECourses with my Coloring Book Flux LoRA. The results were incredibly cute, as Flux developed a unique style that I decided to pursue further. I selected approximately 71 synthetic images and used them to train a LoRA focused on the specific style I wanted to achieve.

FOR THE BEST RESULTS USE EULER OR DEIS SAMPLER

This LoRA works best with objects and human subjects, animals are hit or miss, not enough data in the training images. Something interesting I found is that this actually helps increase the quality in realistic 3D renders of interiors as well.

v1 is a simpler style but is a little overfit, it was trained on 71 images, with 1 repeat, 25 epochs, 32 DIM / 32 ALPHA — 2,486 steps

v2 has better prompt adherence and is much less overfit, it was trained on 101 images, with 1 repeat, 10 epochs, 32 DIM / 32 ALPHA — 1,010 steps  

Trigger keyword: ‘t0yb0x’ ‘simple toy design’ ‘detailed toy design’ ‘3D render’
Recommended strengths: 0.7 – 0.9

Link to Toy Box Flux LoRA on CivitAI
Link to Toy Box Flux LoRA on Hugging Face
Use Toy Box Flux LoRA on Glif