GRAIL is a fully digital data-generation pipeline for humanoid loco-manipulation. It composes 3D assets, simulator-ready scenes, robot-proportioned characters, and video foundation model priors to synthesize metric 4D human-object interaction (HOI) trajectories, then retargets them to a Unitree G1 and trains task-general policies for pick-up, whole-body manipulation, sitting, and terrain traversal. Using only GRAIL-generated data, the resulting egocentric visual policies transfer to real-world object pick-up and stair-climbing.

News

• [2026-08-25] 🎬 MiniMax-H3 video generation — added an optional local MiniMax-H3/Sol-Engine backend for the 2D-HOI pipeline. See the 2D-HOI generation guide.
• [2026-07-15] 🤖 Task-general tracking checkpoints — released policy checkpoints trained on the released GRAIL data. See the tracking guide and dataset page.
• [2026-06-03] 🎉 Code release — released the GRAIL codebase and documentation.

Motion Gallery

Sim-to-Real Deployment

Rendered Egocentric Views

Quick Start

Choose a Docker image and install local extras inside the bind-mounted checkout:

Both Linux AMD64 images contain runtime environments only. They do not contain the GRAIL/Sana source, GRAIL or MiniMax-H3 checkpoints, API keys, or user data. Keep credentials outside the image and provide them only at runtime.

git clone https://github.com/NVlabs/GRAIL.git
cd GRAIL
git submodule update --init --recursive

# The public Docker Hub images can be pulled without registry credentials.
export GRAIL_IMAGE=docker.io/nvgrail/grail:1.1.1
docker pull "$GRAIL_IMAGE"

docker run --gpus all -it --shm-size=16g \
    -v "$PWD":/workspace/grail \
    "$GRAIL_IMAGE"

# inside the container
cd /workspace/grail
bash scripts/setup/install_env_docker.sh   # validates native extensions, downloads Blender
bash scripts/setup/download_checkpoints.sh # GEM-SMPL / GEM-SOMA / FoundationPose weights
bash scripts/setup/download_comasset.sh --category cordless_drill # quick-start object
source /root/miniconda3/etc/profile.d/conda.sh
conda activate grail
[ -f .env ] && source .env                  # OPENAI_API_KEY, KLING_*, HF_TOKEN

The setup script rebuilds GPU/Python-specific native extensions when needed (nvdiffrast, FoundationPose mycpp) and installs Blender into the mounted checkout.

Run any stage end-to-end. Pipeline stages are package entrypoints; invoke them with python -m grail.pipelines.* rather than project-root wrapper scripts.

# 3D asset generation (procedural terrain or AI-generated objects)
python -m grail.pipelines.gen_terrain --type stairs --num 50 --output_dir data/syn_stairs
conda run -n hunyuan python -m grail.pipelines.gen_3d_assets \
    -i configs/gen_3d/example_objects.yaml -o data/gen_example

# 2D HOI generation (Blender + Kling video)
python -m grail.pipelines.gen_2dhoi --dataset ComAsset --category cordless_drill \
    --character kid --results_dir results --video_model_api kling-ai

# 4D HOI reconstruction
python -m grail.pipelines.recon_4dhoi --dataset ComAsset --category cordless_drill --results_dir results

Full install, dataset, and config notes: see Documentation below.

Documentation

Full documentation can be found at docs (rendered HTML) and markdown sources are linked below.

Getting Started

• Installation
• Quick Start

Pipeline

• 3D Asset Generation
• 2D HOI Generation
• 4D HOI Reconstruction
• Retargeting
• Task General Tracking
• Data Export
• Data Visualization
• Web Visualizer

TODOs

• Release task-general tracking policy checkpoints
• Provide quick-start demo script
• Release GRAIL manipulation dataset

Citation

If you find GRAIL useful in your research, please cite:

@misc{grail2026,
  title         = {GRAIL: Generating Humanoid Loco-Manipulation from 3D Assets and Video Priors},
  author        = {Tianyi Xie and Haotian Zhang and Jinhyung Park and Zi Wang and Bowen Wen and Jiefeng Li and Xueting Li and Qingwei Ben and Haoyang Weng and Yufei Ye and David Minor and Tingwu Wang and Chenfanfu Jiang and Sanja Fidler and Jan Kautz and Linxi Fan and Yuke Zhu and Zhengyi Luo and Umar Iqbal and Ye Yuan},
  year          = {2026},
  eprint        = {2606.05160},
  archivePrefix = {arXiv},
  primaryClass  = {cs.RO},
  doi           = {10.48550/arXiv.2606.05160},
  url           = {https://arxiv.org/abs/2606.05160},
}

License

This project is released under the NVIDIA License; see LICENSE for details. The Work and any derivative works may be used only non-commercially, except by NVIDIA Corporation and its affiliates. Third-party components are subject to their own licenses.