The Computable World Model.
A world model is only actionable when the system can model the physics of it: will this hold, fit, overheat, deflect. Real CAD and simulation are too heavy to run where the body is. CadFuture's claim is that the physics an embodied agent needs is mostly retrievable, not computable. Every query walks one cascade, lookup then closed-form formula then sparse solver then a model only as a last resort, over a single graph that carries geometry, the physics fields, tolerances, live-twin state, and the agent's own interaction. The same full engineering model runs from a sub-five-milliwatt neuromorphic chip to a workstation; execution adapts, the engineering truth does not.
Geometry an agent can read is only half of it. The other half is geometry an agent can author, which is what this bench lets you drive.
cad-future on GitHub ↗Illustrative. Physics queries stream in; most resolve at the LUT shelf for about a picojoule, and only a trickle fall through to a solver or model. The research effort is CadFuture, a Charlot Lab project.
↓ Whitepaper · PDFRead online◆ Living paperTechnical Report TR-2026-11 · Institute for Physical AI @ JBI
Geometry an agent can read, simulate and manufacture from works end to end today. Accuracy is lost at one specific join: between the simulator's parameters and the real machine's. That makes calibration the live problem rather than representation, and calibration is the kind of problem that yields to a good instrument.
One of eight, and only one of them is physics. How we read a frontier →