The best ChatGPT that $100 can buy.

Financial datasets for LLMs 🧪

A large scale camera-taken table detection and recognition dataset.

MTL-TabNet: Multi-task Learning based Model for Image-based Table Recognition

Robotics research demonstrating reliability and robustness in the real world (continuously updated)

A collection of tabletop tasks in Mujoco

(RSS 2025) Flow Matching Ergodic Coverage

GPU-friendly, auto-differentiable LQR solver with JAX.

[RAL 2025] MTIL: Encoding Full History with Mamba for Temporal Imitation Learning

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)

[NeurIPS 2023] Efficient Diffusion Policy

A curated list of Diffusion Model in RL resources (continually updated)

RoboDual: Dual-System for Robotic Manipulation

A software framework integrating various imitation learning methods and benchmark environments for robotic manipulation

Unified Implementations of Offline Reinforcement Learning Algorithms

Official Reporsitory of "RoboEngine: Plug-and-Play Robot Data Augmentation with Semantic Robot Segmentation and Background Generation"

Clean implementation of conditional and unconditional behavior transformer.

CLI interfaces & config objects, from types

[ICLR 2024] Official implementation for "Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations"

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

An offline deep reinforcement learning library

A curated list of Decision Transformer resources (continually updated)

High-quality single-file implementations of SOTA Offline and Offline-to-Online RL algorithms: AWAC, BC, CQL, DT, EDAC, IQL, SAC-N, TD3+BC, LB-SAC, SPOT, Cal-QL, ReBRAC

Python library for solving reinforcement learning (RL) problems using generative models (e.g. Diffusion Models).

Implementation of all RL algorithms in a simpler way

FurnitureBench: Real-World Furniture Assembly Benchmark (RSS 2023)

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

[ICLR 2025] Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learning.

[ICRA 2025] Persistent Object Gaussian Splat (POGS) for Tracking Human and Robot Manipulation of Irregularly Shaped Objects