Policies
I want to train a robot
Find policies, datasets, and simulation environments that work with your hardware
25 policies
ACT/AlohaTransferCube
Verified on real hardwareManipulationtransfers a cube from one arm to the other using bimanual coordination
ACT/AlohaInsertion
Verified on real hardwareManipulationinserts a peg into a socket using coordinated bimanual manipulation
DiffusionPolicy/PushT
Verified on real hardwareManipulationpushes a T-shaped block to a target pose on a tabletop
DiffusionPolicy/RobomimicCan
Verified on real hardwareManipulationpicks a can from the table and places it in a target bin
Octo-Base
Verified on real hardwareManipulationexecutes diverse manipulation tasks conditioned on language or goal images across multiple robot embodiments
Octo-Small
Verified on real hardwareManipulationexecutes manipulation tasks with a lightweight generalist model suitable for fine-tuning
pi0-Base
Verified on real hardwareManipulationperforms diverse dexterous manipulation tasks including folding laundry, busing tables, and assembling boxes
RT-2-X
Verified on real hardwareManipulationfollows natural language instructions to manipulate objects on a tabletop across different robot embodiments
LeRobot/DiffusionPushT
Verified on real hardwareManipulationpushes a T-shaped block to a target pose using learned diffusion-based action generation
VINN/FrankaPickPlace
Verified on real hardwareManipulationpicks and places objects by retrieving nearest-neighbor actions from a visual demonstration library
WalkTheseWays/Go2
Verified on real hardwareLocomotionlocomotes with diverse gait styles (walk, trot, bound, pronk) controlled via a continuous gait parameter space
ExtremeParkour/Go2
Verified on real hardwareLocomotionperforms parkour maneuvers including climbing, gap jumping, and leaping over obstacles
MobileALOHA/Cooking
Verified on real hardwareManipulationperforms mobile kitchen tasks like sauteing and wiping while navigating on a wheeled base
ACT/SO100PickPlace
Verified on real hardwareManipulationpicks small objects from a table and places them at a target location
ACT/AlohaClothFolding
Verified on real hardwareManipulationfolds rectangular towels and T-shirts using coordinated bimanual manipulation
RT-1
Verified on real hardwareManipulationfollows natural language instructions for tabletop pick-and-place and drawer manipulation
LeRobot/TDMPC-SimXArm
Tested in simulationManipulationlifts an object using model-predictive control with a learned world model in the simulated xArm environment
H1-BipedalLocomotion
Tested in simulationLocomotionwalks forward on flat ground at commanded velocities up to 1.5 m/s
DiffusionPolicy/KochPickPlace
Tested in simulationManipulationpicks small cubes and places them into a target cup
G1-WholeBodyControl
Tested in simulationManipulationwalks to objects and manipulates them using coordinated whole-body motion
Stretch/HomeNavigation
Tested in simulationNavigationnavigates through home environments to reach specified goal positions while avoiding furniture
BC-RNN/RobomimicSquare
Tested in simulationManipulationplaces a square nut onto the correct peg in simulation
Tello/VisualLanding
Model card onlyAerialdetects an ArUco marker and lands precisely on it using visual servoing
CrossEmbodiment/LocoGo2H1
Model card onlyLocomotionwalks forward using a unified policy across quadruped and humanoid morphologies
ACT/SO100StackCubes
Model card onlyManipulationattempts to stack 2 colored cubes on a tabletop