One robot model, roughly 9,000 different hands: Generalist’s GEN-1 transfers across end effectors

Generalist said on 24 July 2026 that its GEN-1 embodied foundation model now drives about 9,000 end-effector variations — five-fingered hands, two-finger grippers, suction pads, and ordinary tools including power screwdrivers, box cutters, tongs and vegetable peelers — from one set of learned sensorimotor representations rather than a model retrained per gripper. The company says GEN-1 was trained on more than half a million hours of real interaction data. It published no comparative benchmark figures.

What it means Retraining per gripper is the tax that keeps robot deployments bespoke; a model that transfers across hardware is what turns a pilot into a fleet.