Comprehensive suite for evaluating perturbation prediction models

Code for the Molmo Vision-Language Model

robosuite: A Modular Simulation Framework and Benchmark for Robot Learning

robomimic: A Modular Framework for Robot Learning from Demonstration

dataloading is my passion

Efficiently apply modification functions to RLDS/TFDS datasets.

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)

Benchmarking Knowledge Transfer in Lifelong Robot Learning

Octo is a transformer-based robot policy trained on a diverse mix of 800k robot trajectories.

OpenVLA: An open-source vision-language-action model for robotic manipulation.

Collections of robotics environments geared towards benchmarking multi-task and meta reinforcement learning

Codebase for PRISE: Learning Temporal Action Abstractions as a Sequence Compression Problem

CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks