Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning

Viktor Makoviychuk, Lukasz Wawrzyniak, Yunrong Guo, Michelle Lu, Kier Storey +6 more
8/24/2021
cs.ROcs.LG

Official code: Denys88/rl_games (1,400 stars, MIT, PyTorch, last commit 2026-10-19).

Abstract

Isaac Gym offers a high performance learning platform to train policies for wide variety of robotics tasks directly on GPU. Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks. We host the results and videos at \url{https://sites.google.com/view/isaacgym-nvidia} and isaac gym can be downloaded at \url{https://developer.nvidia.com/isaac-gym}.

View on arXivView PDF

Code Implementations(1)

Denys88/rl_gamesOfficial100%
PyTorchMIT

RL implementations

1,400222Python, Jupyter NotebookJan 13, 20196 days agoMIT
deep-learningpytorchreinforcement-learning

Similar Papers

Towards Reinforcement Learning for Space Robotics: Fast Training and Reliable Real-World Transfer

80% match2026

ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI

80% match1 repo3.5k2024

ComSim: Building Scalable Real-World Robot Data Generation via Compositional Simulation

80% match2026

Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments

80% match2 repos8.4k2023

MARBLER: An Open Platform for Standardized Evaluation of Multi-Robot Reinforcement Learning Algorithms

79% match1 repo142023

Habitat: A Platform for Embodied AI Research

79% match1 repo3.8k2019

Cite this paper

@article{makoviychuk2021isaac,
  title  = {Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning},
  author = {Viktor Makoviychuk and Lukasz Wawrzyniak and Yunrong Guo and Michelle Lu and Kier Storey and Miles Macklin and David Hoeller and Nikita Rudin and Arthur Allshire and Ankur Handa and Gavriel State},
  year   = {2021},
  eprint = {2108.10470},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2108.10470}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2108.10470.svg)](https://codeofpaper.com/paper/2108.10470)

Discussion