The RP1 arrives as a modular solution for labs struggling with the limitations of proprietary hardware. Standing 1.3 meters tall, the standard model offers 24 actuated degrees of freedom, while a Pro version expands this to 30 to accommodate complex end-to-end policy testing and human-robot interaction studies. Because the system includes full access to RoboParty’s proprietary ROMOMO joint actuator modules and low-level drivers, developers can now modify the physical execution layer to debug failures that occur between model policy and mechanical response.

This release builds on the momentum of the company's RPO project, which has already attracted significant attention from the developer community. By providing an integrated toolchain—including the PartyOS embodied AI system and the UFO unsupervised reinforcement learning framework—RoboParty is positioning the RP1 as a standardized foundation for whole-body intelligence. According to founder Yi Huang, the goal is to shift innovation away from restricted, siloed environments toward a shared ecosystem where researchers can reproduce results and iterate on both software and physical components simultaneously.