Isaac Sim (pip) and Isaac Lab (Source) Installation

This note records a reproducible installation workflow for Isaac Sim and Isaac Lab, where:

  • Isaac Sim is installed via pip
  • Isaac Lab is installed via git clone

Versions used in this note:

  • Isaac Sim 5.1.0
  • Isaac Lab 2.3.2
  • Python 3.11

Reference documentation:

1. Create the Python environment

First create and activate a new Conda environment. The Python version must match the Isaac Sim version. This note uses Python 3.11:

conda create -n env_isaaclab python=3.11
conda activate env_isaaclab

Then update the system package index and upgrade pip:

sudo apt update
pip install --upgrade pip

2. Install PyTorch

Install the CUDA version of PyTorch specified by the official documentation. The following command is for cu128, which applies to an RTX 5090 setup:

pip install torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cu128

3. Install Isaac Sim via pip

Run the following command to install Isaac Sim:

pip install "isaacsim[all,extscache]==5.1.0" --extra-index-url https://pypi.nvidia.com

After installation, run the following command to verify it:

isaacsim

On first launch, you will be prompted to accept the NVIDIA Omniverse license agreement. Enter Yes in the terminal:

By installing or using Isaac Sim, I agree to the terms of NVIDIA OMNIVERSE LICENSE AGREEMENT (EULA) in https://docs.isaacsim.omniverse.nvidia.com/latest/common/NVIDIA_Omniverse_License_Agreement.html
Do you accept the EULA? (Yes/No): Yes

4. Install Isaac Lab via git clone

Important: the versions of Isaac Sim and Isaac Lab must match. This note uses Isaac Sim 5.1.0 and Isaac Lab 2.3.2.

version_match

Clone the Isaac Lab repository in your working directory. In practice, you may choose the corresponding version according to the Release page:

git clone https://github.com/isaac-sim/IsaacLab.git
cd IsaacLab

If needed, you can inspect the helper script usage with:

./isaaclab.sh --help

Then install the required dependencies for Isaac Lab and run the installer:

sudo apt install cmake build-essential
./isaaclab.sh --install

5. Verify the installation

After installation, you can first run the official minimal example:

./isaaclab.sh -p scripts/tutorials/00_sim/create_empty.py

You can also run it directly with Python in the current virtual environment:

python scripts/tutorials/00_sim/create_empty.py

If you want to further verify the training pipeline, run the official reinforcement learning example:

./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task=Isaac-Ant-v0 --headless

If the commands run normally, the simulator should launch and display a black viewport window. If the window opens correctly, the base installation of Isaac Sim and Isaac Lab is complete.

6. Notes

  • The first launch of isaacsim may be slow, which is usually related to cache initialization and extension loading.
  • The versions of Isaac Sim and Isaac Lab must remain compatible. Before upgrading either component, it is recommended to verify the official compatibility information first.