AI & ML Open Release

Kamino is a massively parallel GPU physics solver that natively supports complex kinematic loops and multi-body systems.

arXiv · March 18, 2026 · 2603.16536

Vassilios Tsounis, Guirec Maloisel, Christian Schumacher, Ruben Grandia, Agon Serifi, David Müller, Chris Amevor, Tobias Widmer, Moritz Bächer

The Takeaway

Standard robotics simulators (like MuJoCo) often struggle with closed-loop kinematics, requiring practitioners to use simplified 'tree' approximations. Kamino allows high-throughput RL training on robots with complex mechanical topologies directly on the GPU.

From the abstract

We present Kamino, a GPU-based physics solver for massively parallel simulations of heterogeneous highly-coupled mechanical systems. Implemented in Python using NVIDIA Warp and integrated into the Newton framework, it enables the application of data-driven methods, such as large-scale reinforcement learning, to complex robotic systems that exhibit strongly coupled kinematic and dynamic constraints such as kinematic loops. The latter are often circumvented by practitioners; approximating the syst