We're accelerating the timeline to a world with billions of robots, and making sure they're accessible, auditable, and beneficial to humanity.
Join the RL training challengeGeneral-purpose humanoid robots for developers, hobbyists, and researchers
Our software, hardware, and machine learning stack is seemlessly integrated, allowing you to focus on building applications instead of installing packages.
Owning every layer—from metal to model—lets us move faster, integrate more deeply, and open-source each breakthrough for the whole world to build on.
Experience our humanoid robots in action
Create, deploy, and share robot apps, tools, and policies—built by you and the community.
A single hub to install apps, customize behaviours, train new skills, and teleoperate your robot.
Control your robot with a neural interpretation domain-specific language (DSL).
Open-source library for GPU-accelerated robot learning and zero-shot sim-to-real transfer
High-performance reinforcement learning framework optimized for training humanoid robot locomotion, manipulation, and real-world deployment.
High versatility for tasks such as walking, dancing, household organization, and even cooking.
A generalist policy using large-scale robot data with novel network architecture to enable the most capable and dexterous robot
Locally run and capable of integrating with other VLAs such as Pi0.5 and Gr00t.
Rust-based OS to run policies on the real robot, or evaluate in simulation
Rust-based, fast, and reliable robot operating system combining hardware, software, and firmware.
Ships with an easy-to-use Python SDK to easily develop robot applications.
A digital twin / model evaluator for the K-Scale Operating System (K-OS) using the same gRPC interface as our real robot
Effortlessly test and refine your models in simulation
Deploy policies, VLA models and applications on robot hardware in real-time
A lightweight, modular framework for developing reinforcement learning policies in simulation and subsequently deploying on physical robots
Train policies in under an hour. Leveraging GPU accelerated libraries like JAX and Mujoco XLA, K-Sim can process over 100,000 samples per second (including training) on an RTX 4090.
Quickly pinpoint the root cause of training regressions or improvements. All of the code defining experiment lives neatly in a single file which is automatically logged to TensorBoard.
A trained policy is immediately ready to be exported and deployed on a real robot, saving countless hours otherwise spent on maintaining two sets of infrastructure for training and deployment.
K-Sim can be easily modified to fit your use-case, instead of the other way around. All of the code is fully open source so you can build on top of it with confidence.
| K-Sim | Issac | Brax | Gazebo | |
|---|---|---|---|---|
| Differentiable physics | ||||
| GPU-accelerated | ||||
| TPU-accelerated | ||||
| Cross-platform | ||||
| Single-file training scripts | ||||
| Officially supported hardware | ||||
| Motion capture priors | ||||
| Stateful (RNNs, Transformers) |
We've completed 6 generations of robots in less than a year
Collaborate, learn, and innovate with us
Get rapid-fire development support through our Discord—home to 2000+ active members who have collaborated on 6 humanoid robots and counting.
We're engineers, hackers, and researchers that believe in a world where robots are made for everyone.
Shoot us an email at for any business inquiries or if you want to connect with us!
Backed by world-class investors