Kineforge trains embodied control policies without human teleoperation demos. Frozen semantic priors + differentiable physics (MuJoCo MJX) + a 157K-parameter policy = training in hours on a single GPU, and CPU inference …
Kineforge trains embodied control policies without human teleoperation demos. Frozen semantic priors + differentiable physics (MuJoCo MJX) + a 157K-parameter policy = training in hours on a single GPU, and CPU inference under 10ms. Structure over scale.
讨论这个产品