Models

Research

Repository for our papers: Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics and Uncertainty-Aware Robotic World Model Makes Offline Model-Based…

Open source

Code accompanying the paper "Learning Agile Robotic Locomotion Skills by Imitating Animals"

Open source

SPA: 3D Spatial-Awareness Enables Effective Embodied Representation

Research

Robot policy research using latent action representations.

Research

Robot manipulation model resources from the Allen Institute for AI.

Open source

Official codebase for Manipulation Primitive-augmented reinforcement Learning (MAPLE)

Research

InternNav provides a platform for developing navigation foundation models.

Research

DayDreamer provides world-model methods for physical robot learning.

Research

A video-action model research project for robotic control.

Open source

DexMV: Imitation Learning for Dexterous Manipulation from Human Videos, ECCV 2022

Research

Object-pose estimation research tools for robotic perception.

Research

Implementation of π₀, the robotic foundation model architecture proposed by Physical Intelligence

Research

GraspVLA provides grasping-model research pretrained with synthetic action data.

Research

Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies

Open source

Source codes for the paper "3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning"

Open source

Official repository of T-Rex: Tactile-Reactive Dexterous Manipulation

Research

🔥 SpatialVLA: a spatial-enhanced vision-language-action model that is trained on 1.1 Million real robot episodes. Accepted at RSS 2025.

Research

official implementation for our paper Steering Your Generalists: Improving Robotic Foundation Models via Value Guidance (CoRL 2024)

Open source

Language-guided whole-body control for general humanoid robots

Open source

Official Code for "Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents"

Research

Official code for MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation

Research

BAAI UniVLA provides a unified vision-language-action model research project.

Open source

"FALCON: Learning Force-Adaptive Humanoid Loco-Manipulation"

Research

Official implementation of the paper: "PhysHSI: Towards a Real-World Generalizable and Natural Humanoid-Scene Interaction System"