Last updated: 2026-03-18 · 作者:@qclawhq· 来源:Hugging Face 官方仓库 + 社区实测

This site is an independent community hub, not affiliated with Hugging Face. Official repo: https://github.com/huggingface/lerobot

How to Choose LeRobot Models and Datasets for Fast Reproduction?

Direct Answer: Filter models by hardware first, then choose the algorithm type (Diffusion/ACT/TD-MPC2) best suited for your task.

AlgorithmProsConsBest For
Diffusion PolicySmoother actionsHeavy inferenceFine grasping
ACTStable to trainData sensitiveDual-arm tasks
TD-MPC2Strong dynamic controlComplex tuningHumanoids/Walking
HardwareRecommended Use
SO-100Beginner grasp loop
AlohaFine dual-arm operation
UnitreeDynamic tasks & mobility
Reproduction MethodProsRisksRecommendation
Download & InferenceFastestEnvironment diffValidate motion first
Fine-tuneFast convergenceTransfer failureSimilar tasks
Train from scratchMost controllableHigh costVery different tasks

lerobot/diffusion_pusht

Diffusion Policy

Diffusion Policy model trained in the PushT environment (gym-pusht) to push a T-shaped block to a target position. Official reference implementation of Diffusion Policy in LeRobot.

Hardware: PushT / Simulation 128 stars 1.2k downloads
HF Link

lerobot/act_aloha_sim_transfer_cube_human

ACT

ACT (Action Chunking with Transformers) policy trained in the Aloha simulation environment for cube transfer tasks, using human teleoperation demonstrations.

Hardware: Aloha (Simulation) 456 stars 5.4k downloads
HF Link

lerobot/act_aloha_sim_insertion_human

ACT

ACT policy trained in the Aloha simulation environment for fine-grained insertion tasks (AlohaInsertion), using human teleoperation demonstration data.

Hardware: Aloha (Simulation) 89 stars 800 downloads
HF Link

Want to Share Your Model?

If you have trained an excellent model or compiled a high-quality dataset, welcome to submit it to us. Authors featured on the curated list will receive a special lerobot.tech community badge and hardware discounts.

Submit Model Card