zengxianyu
/

ppd

Model card Files Files and versions

Models for this repo and this paper arxiv.org/abs/2512.05106

The 4-step lora checkpoints for Wan2.2-14b are converted from Wan2.2-Lightning

Lora for FLUX.1-dev: flux1-dev_phipd_lora_302000.safetensors

Lora for Wan2.2-14b High noise: wan2.2-14b-high-step-12400.safetensors

Low noise wan2.2-14b-low-step-12400.safetensors

SD1.5 sd1.5-epoch=10-step=720000.ckpt

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for zengxianyu/ppd

Base model

Wan-AI/Wan2.2-I2V-A14B

Finetuned

(45)

this model

Paper for zengxianyu/ppd

NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation

Paper • 2512.05106 • Published Dec 4, 2025 • 16