chiawenchen23/nap-control-checkpoints

Official model checkpoints for NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control

---

🚀 Quick Start & Download

Option 1: Automated Download (Recommended)

The easiest way to get started is to use the setup script provided in the main GitHub repository. Follow the README.md and run:

bash download_data.sh

This script will automatically fetch all checkpoints from Hugging Face and place them into their correct directory paths.

Option 2: Manual Download & Organization

If you prefer to download the files manually, you must arrange them into the following directory structure inside your workspace root:

|-- assets
|-- data
|-- nap
|-- output
    |-- HumanoidIm
        |-- agile_goal
            |-- Humanoid.pth
        |-- agile_goal_terrain
            |-- Humanoid.pth
        |-- far_goal
            |-- Humanoid.pth
        |-- far_goal_terrain
            |-- Humanoid.pth
        |-- multi_goal
            |-- Humanoid.pth
        |-- pulse_vae_iclr
        |-- sit
            |-- Humanoid.pth
        |-- traj
            |-- Humanoid.pth
        |-- velocity
            |-- Humanoid.pth
        |-- velocity_terrain
            |-- Humanoid.pth
...
|-- UniPhys
    ...
    |-- isaac_utils
    |-- output
        |-- HumanoidIm
        |-- root_with_dof
            |-- checkpoints
                |-- last.ckpt
            |-- train_data_stats.npy
...

📜 Citation

If you find our work, code, or checkpoints useful for your research, please cite our paper:

@misc{chen2026napcontrolnavigatingdiffusionprior,
      title={NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control},
      author={Chia-Wen Chen and Yan Wu and Korrawe Karunratanakul and Siyu Tang},
      year={2026},
      eprint={2605.20209},
      archivePrefix={arXiv},
      primaryClass={cs.GR},
      url={https://arxiv.org/abs/2605.20209},
}