Eka raises $150 million for force-aware robot manipulation

Eka Robotics has raised a $150 million Series A at a reported $1.45 billion valuation, backing an approach to robotic manipulation that combines visual information with feedback from physical contact. The financing was reported on October 20, as lead investor Andreessen Horowitz published its investment announcement.

In that primary statement, the investor describes a Vision-Force-Action model, with sensing, control and hardware designed together. The aim is to let a robot respond when a grasp slips or a connector resists insertion, rather than relying entirely on what a camera sees before contact.

That is an important practical distinction. Seeing where an object is does not settle how firmly to hold it, whether it is deforming, or whether a part has caught on an edge. A system that can adjust after touching an object could be useful in work where materials and positions vary. This remains an engineering proposition whose value depends on reliable execution.

Eka's own overview presents the technology as a model that can operate across objects, tasks and environments. Those are company claims, not an independently established description of unrestricted capability. Its broad language should not be read as proof that the system can perform any industrial job safely.

The investor says Eka trains manipulation policies through reinforcement learning in physics simulation. It also acknowledges the difficulty of transferring simulated contact dynamics to physical hardware. Friction, compliance and actuator behavior can differ between a virtual environment and a production line, making successful demonstrations an incomplete test of deployment readiness.

SiliconANGLE independently reported the financing, identifying Andreessen Horowitz and F-Prime as leaders of the round. Its account also distinguishes company-reported picking performance from unresolved questions about running reliably over long periods and extending the approach beyond the demonstrated tasks.

The investment provides resources to pursue those questions; the valuation does not answer them. Buyers would still need evidence about their own objects, failure recovery, maintenance requirements and sustained operation. The development is a funding and technology announcement, with the harder commercial test still taking place outside the demonstration setting.