Cosmos is a world model development platform that consists of world foundation models, tokenizers and video processing pipeline to accelerate the development of Physical AI at Robotics & AV labs. C…

World's First Large-scale High-quality Robotic Manipulation Benchmark

[CoRL 2024] HumanPlus: Humanoid Shadowing and Imitation from Humans

SAPIEN Manipulation Skill Framework, a open source GPU parallelized robotics simulator and benchmark, led by Hillbot, Inc.

Evaluating and reproducing real-world robot manipulation policies (e.g., RT-1, RT-1-X, Octo) in simulation under common setups (e.g., Google Robot, WidowX+Bridge) (CoRL 2024)

Official implementation of "Data Scaling Laws in Imitation Learning for Robotic Manipulation"

A generative world for general-purpose robotics & embodied AI learning.

Digit 360 is a modular platform that unlocks new capabilities, and enables future research on the nature of touch.

SAPIEN Embodied AI Platform

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

[ICLR 2025] SPA: 3D Spatial-Awareness Enables Effective Embodied Representation

[NeurIPS 2024] Behavioral Topology (BeTop), a multi-agent behavior formulation for interactive motion prediction and planning

RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins

[ECCV 2024] Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting

[NeurIPS 2024] CLOVER: Closed-Loop Visuomotor Control with Generative Expectation for Robotic Manipulation

Generative model for Unified Motion Planning tasks

Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimality (NeurIPS 2023, Spotlight)

Mirage: a zero-shot cross-embodiment policy transfer method. Benchmarking code for cross-embodiment policy transfer.

A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.

world modeling challenge for humanoid robots

Codebase for the WayveScenes101 Dataset

BEVFormer, UniAD, VAD in Closed-Loop CARLA Evaluation with World Model RL Expert Think2Drive

[NeurIPS 2024 Datasets and Benchmarks Track] Closed-Loop E2E-AD Benchmark Enhanced by World Model RL Expert

[CVPR'24 Best Student Paper] Mip-Splatting: Alias-free 3D Gaussian Splatting

[NeurIPS 2024] A Generalizable World Model for Autonomous Driving

[RSS 2024] Learning Manipulation by Predicting Interaction

Multi Agent Traffic Scenario Gym: A scenario-based training and evaluation framework for CARLA.