kinder-bench/kinder-DP-checkpoints

KinDER — Diffusion Policy (DP) Checkpoints

Trained Diffusion Policy checkpoints for the KinDER physical-reasoning benchmark (RSS 2026).

Each checkpoint is an imitation learning policy trained from ~100 human demonstrations per environment using RGB image observations. The training code lives in kinder-diffusion-policy (a fork of diffusion_policy). Demonstrations are available at kinder-bench/kinder-datasets.

---

Checkpoints

Path	KinDER environment	Trained epochs	Final train loss
motion2d/epoch=1000-train_loss=0.000.ckpt	Motion2D-p0	1 000	0.000
stickbutton2d/epoch=2000-train_loss=0.001.ckpt	StickButton2D-b1	2 000	0.001
dynobstruction2d/epoch=2000-train_loss=0.000.ckpt	DynObstruction2D-o1	2 000	0.000
dynpushpullhook2d/epoch=0900-train_loss=0.001.ckpt	DynPushPullHook2D-o5	900	0.001
basemotion3d/epoch=2000-train_loss=0.000.ckpt	BaseMotion3D	2 000	0.000
shelf3d/epoch=0300-train_loss=0.000.ckpt	Shelf3D	300	0.000
sweep3d/epoch=0300-train_loss=0.001.ckpt	SweepIntoDrawer3D	300	0.001
transport3d/epoch=0100-train_loss=0.000.ckpt	Transport3D-o2	100	0.000

---

Method

Diffusion Policy (DP) learns a visuomotor policy by training a diffusion model to denoise action sequences conditioned on RGB image observations.

• Input: RGB image(s) at 224 × 224 (single overhead camera for 2D envs; base + wrist + overview cameras for 3D TidyBot envs)
• Output: action chunk predicted by iterative DDPM denoising
• Training data: ~100 human demonstrations per environment (collected via PS5 controller for 2D; iPhone / VR headset for 3D)

---

Usage

1. Install dependencies

# Clone and set up kinder-diffusion-policy
git clone git@github.com:Princeton-Robot-Planning-and-Learning/kinder-diffusion-policy.git
cd kinder-diffusion-policy
# Follow the environment setup instructions in the repo README
mamba activate robodiff

# Install the kinder-imitation-learning inference utilities
cd kinder-baselines/kinder-imitation-learning
uv pip install -r prpl_requirements.txt
uv pip install -e ".[develop]"

2. Launch the policy server

cd ~/kinder-diffusion-policy
mamba activate robodiff
python policy_server.py --ckpt-path /path/to/sweep3d/epoch=0300-train_loss=0.001.ckpt

3. Run evaluation

cd kinder-baselines/kinder-models/scripts
python inference.py \
    --env-name kinder/SweepIntoDrawer3D-o5-v0 \
    --save-videos \
    --num-seeds 1 \
    --num-episodes 5 \
    --max-steps 200

Replace --env-name and --ckpt-path with the environment and checkpoint of your choice.

---

Training from scratch

To retrain a checkpoint on your own data:

# Convert raw teleoperation recordings to HDF5
cd kinder-baselines/kinder-models/scripts
python demos_to_hdf5.py \
    --teleop_data_dir $YOUR_DATA_DIR \
    --output_path $OUTPUT_HDF5_PATH \
    --render_images

# Train (example config for SweepIntoDrawer3D)
cd ~/kinder-diffusion-policy
mamba activate robodiff
python train.py --config-name=train_sweep3d_image

---

Related resources

Resource	Link
KinDER benchmark	kindergarden
Training code	kinder-diffusion-policy
Demonstration datasets	kinder-bench/kinder-datasets
DP + Environment States (DPES) checkpoints	kinder-bench/kinder-DPES-checkpoints
Finetuned π0.5 VLA checkpoints	kinder-openpi

---

Citation

If you use these datasets, please cite the paper: KinDER: A Physical Reasoning Benchmark for Robot Learning and Planning:

@inproceedings{huang2026kinder,
  title     = {KinDER: A Physical Reasoning Benchmark for Robot Learning and Planning},
  author    = {Huang, Yixuan and Li, Bowen and Saxena, Vaibhav and Liang, Yichao and Mishra, Utkarsh and Ji, Liang and Zha, Lihan and Wu, Jimmy and Kumar, Nishanth and Scherer, Sebastian and Xu, Danfei and Silver, Tom},
  booktitle = {Robotics: Science and Systems (RSS)},
  year      = {2026}
}