LoRA extraction works fairly well for stylized image generation.

LoRA extraction works fairly well for stylized image generation.

Trained on SDXL 1.0 for realism.

Extracted with Kohya SS GUI 128x128 rank

Images generated on starlightXLAnimated_v3 model.

Didn't spend much time for prompt optimization. A simple prompt

cartoon animated animation digital artwork image of ohwx man wearing an amazing armor <lora:lora_test_1:1.0>

Negative prompt: photo photograph