GenesisBench

How language intelligence can be used to improve physical intelligence.

How it works

GenesisBench evaluates how LLM agents can train the policies that run on real robot hardware.

Heuristic policies in simulation

No training and no H100s yet — agents iterate on code logic alone, optimizing heuristic policies for performance in simulation.

The reference task runs this improvement loop on the MuJoCo Ant environment (Gymnasium).

Starter policya runnable CPG/PD gait — weaker than the reference
Submit the best policyscored on a hidden Ant suite · starter = 0, reference = 100

Mirrors the simulation-heuristics Ant v1 reference task; snippets are illustrative.

Environments

Two environments: simulation for training and evaluation, the real world for operation.

Environments

Simulation

Policies are trained and evaluated in simulation — the simulator shows how they are really working.

Real-world operation

The goal: policies that can be used to run on real robotics hardware.

real robot hardware

Research background

GenesisBench builds on research at the intersection of agentic coding, post-training, and embodied AI.

ENPIRE: Agentic Robot Policy Self-Improvement in the Real WorldNVIDIA GEAR · arXiv:2606.19980 Xiao, Xie, Zhang, Lin, et al. — LLM agents improving robot policies through closed-loop real-world iteration.
All references11