Models

Research

OpenDW provides an action-conditioned world-model project for embodied intelligence.

Research

Openpi Dexterous studies pi0 LoRA adaptation for Shadow Hand manipulation and reports experimental limitations.

Research

Open Pi Zero is a reimplementation of Physical Intelligence's pi0 VLA model.

Research

OpenVLA Reproduction documents setup and evaluation of OpenVLA on LIBERO tasks.

Research

Official code for "OpenWAM: An Open Framework for Composable World-Action Models" (Stanford University)

Research

O-VLA studies VLA transfer through semantic primitive learning.

Research

PointWorld studies 3D world models for robotic manipulation.

Research

Official Implementation of the ICLR 2024 spotlight paper: Universal Humanoid Motion Representations for Physics-Based Control

Research

Official implementation of ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver.

Research

The code of paper "From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidance" accepted by ICLR'26

Research

A proactive robot manipulation model for multimodal physical environments

Research

Official repo of Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics

Research

We present the first successful deployment of a whole-body sampling-based MPC system on a real-world quadruped robot. Our approach enables real-time, contact-rich locomotion and manipulation using…

Research

RynnValue: Scaling Robotic Value Foundation Models with Temporal Distance

Research

RynnWorld-4D: 4D Embodied World Models for Robotic Manipulation

Research

Research connecting language instructions with executable robot skills.

Research

This is code repo for the paper, Depolying Ten Thousand Robots: Scalable Imitation Learninig for Lifelong Mulit-Agent Path Finding, which won the ICRA 2025 best paper on multi-robot systems and…

Research

The official implementation of the paper "Scaling Behavior Foundation Model for Humanoid Robots"

Research

repository for training action-conditioned latent diffusion world models for robot video generation

Research

Offical implementation of ICLR'26 paper "Empowering Multi-Robot Cooperation via Sequential World Models"

Research

Native-video memory for vision-language-action models, using timestamped visual history and exact streaming inference for long-horizon robot manipulation.

Research

Official implementation of SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation

Research

Spirit-v1.5: A Robotic Foundation Model by Spirit AI

Research

Theia: Distilling Diverse Vision Foundation Models for Robot Learning