Research
Generalist’s latest embodied foundation model, GEN-1, now supports a broad range of robot end effectors, from five-fingered hands to specialized tools. By training GEN-1 to work with these new hands, Generalist is demonstrating that a single base AI model can learn sensorimotor policies that transfer across radically different ways of interacting with the physical world.
GEN-1 is pretrained on Generalist’s in-house robotics dataset, which spans a wide variety of end effectors across more than half a million hours of real interaction data. Some end effectors involve new form factors with their own actuation schemes and camera positions, while others are off-the-shelf tools, printed parts, or custom modifications to the company’s standard two-finger grippers — approximately 9,000 variations so far.
By training GEN-1 to work with these new hands, Generalist is demonstrating that a single base AI model can learn sensorimotor policies that transfer across radically different ways of interacting with the physical world.
Each end effector is a different sensorimotor interface through which GEN-1 experiences the physical world, learning about geometry, contact, friction, forces, and dynamics. Scaling pretraining across thousands of these interfaces teaches GEN-1 universal sensorimotor representations that transfer to new hands and new ways to grasp, push, pull, twist, and more.
Source: The Robot Report