Welcome, RLDX-1 by RLWRLD!
The RLDX-1 is a robotics foundation model constructed by RLWRLD, aimed at fostering advanced robotic manipulation and industrial automation. Rather than focusing merely on basic visual understanding, it emphasizes dexterous hand control. The model integrates vision, language, tactile feedback, force sensing, and memory, thereby facilitating robots to accomplish sophisticated real-world tasks with increased precision and adaptability.
What sets RLDX-1 apart from conventional VLA models is its prominence on physical interaction and long-term task execution. The model caters to humanoid robots, dual-arm systems, and robotic hands with multiple degrees of freedom, addressing the diverse needs of robotic systems. With its design tailored for authentic industrial settings such as manufacturing and logistics, it equips robots to function in varied and complex environments.
New! 2026 Humanoid
Robot Market Report
198 pages of exclusive insight from global robotics experts — uncover funding trends, technology challenges, leading manufacturers, supply chain shifts, and surveys and forecasts on future humanoid applications.
Aaron Saunders, Former CTO of
Boston Dynamics,
now Google DeepMind
RLWRLD released the RLDX-1 model to the public in May 2026 and has endeavored to support and collaborate with the larger robotics community by sharing research resources and checkpoints. Currently, the model is on backorder, indicating its significant demand and acceptance in the industry. The broader implications of its release signal a shift towards more advanced and capable robotic systems, capable of both precise control and complex manipulation.
