Why there are no Krea2 / MOSS versions:
I've decided not to port this LoRA to Krea2 or MOSS. Since these modern architectures natively understand and generate this footwear design out of the box, training a LoRA for it isn't needed—and would only dilute or pollute the base model.
F1D
The dataset is for reference only, the training folder is lost, but all images are from my uploaded dataset. I selected about 4-5 product images and around 20 images of people wearing the products, totaling about 25 images, trained for approximately 2000 steps.
This Lora model has an issue where the shoes often appear blurry. Please try modify parameter tuning or use Inpainting to fix it.
这个 Lora 模型有点问题,鞋子容易出现模糊现象,请反复调参或者 Inpainting 修复
Prompts
She is wearing sneakers, Air Jordan 1, perforated toe box, flat laces, Swoosh logo, nike logo, Wings logo, Jumpman logo on tongue, classic sole, high-top, shoes, nike \(company\),
white toe, red mudguard, red eyelets, red heel, black ankle collar, black swoosh, black laces,Description
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.


