> ## Documentation Index > Fetch the complete documentation index at: https://mintlify.com/huggingface/lerobot/llms.txt > Use this file to discover all available pages before exploring further. # Overview # Simulation Environments LeRobot provides a comprehensive suite of simulation environments for training and evaluating robotic policies. Simulation enables rapid iteration, scalable data collection, and reproducible benchmarking before deploying to real robots. ## Why Simulation? Simulation environments are essential for: * **Rapid prototyping**: Test ideas quickly without hardware constraints * **Scalable data collection**: Generate large datasets efficiently * **Reproducible evaluation**: Benchmark policies with standardized tasks * **Safe exploration**: Train policies without risking physical damage * **Parallel training**: Run multiple environments simultaneously for faster learning ## Available Environments LeRobot integrates several popular simulation platforms: ### GPU-Accelerated Environments * **[LeIsaac](./leisaac)**: IsaacLab-based everyday manipulation tasks with SO101 robots * Teleoperation workflows for data collection * Single-arm and bi-arm tasks * Everyday skills: picking, lifting, cleaning, folding * **[NVIDIA IsaacLab Arena](./isaaclab-arena)**: High-fidelity humanoid manipulation * GR1, G1, and Galileo humanoid robots * RTX rendering for vision-based policies * Massively parallel GPU rollouts ### Benchmark Environments * **[LIBERO](./libero)**: Lifelong learning benchmark with 130 tasks * Five task suites focusing on spatial, object, and goal reasoning * Long-horizon manipulation tasks * Knowledge transfer evaluation * **[MetaWorld](./metaworld)**: Multi-task reinforcement learning benchmark * 50 diverse tabletop manipulation tasks * Standardized difficulty splits (easy, medium, hard) * Generalization evaluation ## Using Environments from the Hub LeRobot's **[EnvHub](./envhub)** provides one-line environment loading from HuggingFace Hub: ```python theme={null} from lerobot.envs.factory import make_env # Load any environment from the Hub envs = make_env( "username/environment-repo", n_envs=4, trust_remote_code=True ) ``` This enables: * Instant environment sharing and reproducibility * Version control for environments * Community-contributed tasks * Zero-setup environment loading Learn more in the [EnvHub guide](./envhub). ## Environment Integration All LeRobot simulation environments follow a unified interface: ```python theme={null} from lerobot.envs.factory import make_env # Create vectorized environments env_dict = make_env( cfg, # Environment configuration n_envs=4, # Number of parallel environments use_async_envs=False # Sync or async vectorization ) # Access environments by suite and task suite_name = next(iter(env_dict)) vec_env = env_dict[suite_name][0] # Standard Gym interface obs, info = vec_env.reset() for _ in range(1000): actions = policy.select_action(obs) obs, rewards, terminated, truncated, info = vec_env.step(actions) ``` ### Observation Structure LeRobot uses a consistent observation format across environments: ```python theme={null} { "observation.images.image": torch.Tensor, # Main camera view "observation.images.image2": torch.Tensor, # Optional second camera "observation.state": torch.Tensor, # Proprioceptive state "task": List[str] # Task descriptions (for VLAs) } ``` ### Action Space Actions are continuous control commands: * **Format**: `torch.Tensor` or `np.ndarray` * **Range**: Typically `[-1, 1]` normalized * **Shape**: Environment-specific (e.g., 7-DoF for LIBERO, 4-DoF for MetaWorld) ## Training with Simulation LeRobot provides integrated training loops that combine simulation environments with policy learning: ```bash theme={null} lerobot-train \ --policy.type=smolvla \ --policy.repo_id=${HF_USER}/my-policy \ --dataset.repo_id=lerobot/libero \ --env.type=libero \ --env.task=libero_10 \ --steps=100000 \ --batch_size=4 \ --eval_freq=1000 ``` Key features: * **Online evaluation**: Automatically evaluate during training * **Multi-suite support**: Train on multiple task suites simultaneously * **Flexible scheduling**: Control evaluation frequency and episode counts * **Automatic checkpointing**: Save best models based on success rate ## Evaluation Benchmark trained policies on simulation environments: ```bash theme={null} lerobot-eval \ --policy.path=lerobot/pi05_libero_finetuned \ --env.type=libero \ --env.task=libero_spatial,libero_object \ --eval.batch_size=2 \ --eval.n_episodes=10 ``` Evaluation features: * **Multi-task evaluation**: Test across multiple tasks/suites * **Parallel rollouts**: Run multiple environments simultaneously * **Success metrics**: Automatic success rate computation * **Video recording**: Save rollout videos for analysis ## Environment Configuration Each environment can be configured through dataclass configs: ```python theme={null} from lerobot.envs.configs import LiberoEnv, MetaworldEnv # LIBERO configuration libero_cfg = LiberoEnv( task="libero_10", task_ids=[0, 1, 2], # Specific tasks episode_length=500, # Max steps per episode obs_type="pixels_agent_pos", # Observation type camera_name="agentview_image", # Camera selection control_mode="relative" # Control parameterization ) # MetaWorld configuration metaworld_cfg = MetaworldEnv( task="medium", # Difficulty level or task list obs_type="pixels", # Observation type episode_length=400 # Max steps per episode ) ``` ## Performance Tips ### GPU Acceleration For IsaacLab-based environments (LeIsaac, Arena): * Use `--eval.batch_size` to control parallel environments * Enable RTX rendering for vision-based policies * Run headless for maximum throughput ### CPU Environments For LIBERO and MetaWorld: * Set `MUJOCO_GL=egl` for headless rendering * Use `--eval.batch_size` for parallel rollouts * Consider `AsyncVectorEnv` for better CPU utilization ### Memory Management * Reduce `observation_width` and `observation_height` to save memory * Limit `n_envs` based on available RAM/VRAM * Use lower batch sizes for evaluation on limited hardware ## Next Steps Load and share environments from HuggingFace Hub Control SO101 robots in IsaacLab simulation GPU-accelerated humanoid manipulation Lifelong learning benchmark Multi-task RL benchmark ## Contributing Environments We welcome new simulation environments! To contribute: 1. **Implement the environment** following Gym interface 2. **Create an EnvConfig** in `lerobot/envs/configs.py` 3. **Add factory logic** in `lerobot/envs/factory.py` 4. **Upload to EnvHub** for easy sharing 5. **Submit a PR** with documentation See our [contribution guide](https://github.com/huggingface/lerobot/blob/main/CONTRIBUTING.md) for details.