PDMD 4-NFE LoRA for MiniMax-H3

Paper (arXiv) · Project page

This repository holds the LoRA adapter of a 4-step (4 NFE) student distilled from MiniMax-H3 with Projected Distribution Matching Distillation (PDMD). The same model is also available as full transformer weights in pdmd2026/pdmd_4NFE_full.

The adapter is applied to the transformer (MiniMaxH3Transformer3DModel) of the base model. Every other component (VAE, audio VAE, schedulers, text encoder, processor) is unchanged and comes from the base model.

Files

Tensor keys have the form transformer.<module>.lora_A.weight / transformer.<module>.lora_B.weight, where <module>.weight is the corresponding parameter of MiniMaxH3Transformer3DModel. Fusing rule (also stored in the safetensors metadata):

W_base += lora_scale * (lora_B @ lora_A)      # lora_scale = 1.0 (alpha / rank = 128 / 128)

Sample with 4 denoising steps using the base model's released scheduler configuration (shift 12 / 3).

Citation

@misc{wang2026pdmdprojecteddistributionmatching,
      title={PDMD: Projected Distribution Matching Distillation for Video Diffusion Models},
      author={Zimo Wang and Junkun Yuan and Angtian Wang and Haotian Yang and Canyu Zhang and Siyuan Yuan and Xingchang Huang and Bo Liu and Yizhi Wang and Yiding Yang and Chongyang Ma and Gordon Guocheng Qian},
      year={2026},
      eprint={2609.35768},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2609.35768},
}