Rimnot, a physical AI startup founded by Tsinghua University graduates, has partnered with Yuantu Future Technology. The collaboration aims to implement 10,000 industrial robots for server manufacturing needs, marking one of the largest commercial steps in the embodied AI industry.
At the WAIC exhibition, Rimnot demonstrated its capabilities by completing a real-time server assembly task. The robot handled motherboard components, including densely packed chips, interfaces, and high-precision parts. This required recognizing various model variants, as well as precise control over gripping force and placement.
Despite the limited equipment uptime at the exhibition, the robot managed to achieve full environmental adaptation in just 30 minutes of training. It showed the system's ability to generalize knowledge across different conditions without extensive reprogramming, even when lighting, temperature, and spatial layout changed.
The core of the technology is Rimnot's fundamental physical world model combined with a robotic system that ensures a complete closed loop—from process understanding and task planning to execution feedback and continuous optimization. Unlike digital AI, where one model can serve millions of users, physical AI requires robots to operate in specific environments, process defined objects, and perform designated actions while accounting for visual errors, force feedback, and unforeseen real-world situations that cannot be fully simulated in data centers.
Yuantu Future Technology brings production scale and deep industry expertise to the project. As a global leader in server manufacturing and full-stack AI infrastructure solutions, Yuantu produces AI servers, general servers, switches, supercomputers, and liquid-cooled racks, offering a complete portfolio from R&D to service.
Server manufacturing is viewed as an ideal scenario for the commercialization of embodied AI because it involves complex tasks such as assembly, cable connection, functional testing, and quality control, possessing sufficient industrial value to justify large-scale robot deployment. The partnership aims to transition from replacing individual processes to full closed-loop robotic operations. Rimnot provides the physical world models and industrial solutions, while Yuantu offers real production scenarios and operational experience.
If successfully implemented at the 10,000 unit level across Yuantu's production sites, this collaboration could create a reproducible model for deploying embodied AI in precision electronics manufacturing, opening up a market estimated by Rimnot to be hundreds of billions of yuan. The rapid pace from company foundation to a mass deployment commitment in just four months suggests that the bottleneck in physical AI is not technology development, but the realization of industrial partnerships.
Dexmal has introduced a comprehensive suite of products for embodied artificial intelligence aimed at overcoming the final engineering barrier between AI models and real-world performance. Among the announced developments are the universal foundational model DM0.5, the Apex universal robot, the DexOS operating system, and the MaaS platform.
The company's strategy is built on a three-stage model, analogous to a rocket: the first stage focuses on creating a generalized foundational model; the second stage includes developer tools such as MaaS and the OS; and the third stage concentrates on systematic scenario solving capabilities.
The DM0.5 foundational model possesses 4 billion parameters and was trained on 150,000 hours of multi-source data. This represents a fourfold increase in data and a twofold increase in parameters compared to the DM0 model. The model demonstrates a low inference latency of 50 ms, a 25% improvement in inference efficiency, and achieved first place in the RoboChallenge Table30 V2 ranking with an overall score of 60.1, while showing 99.1% complex performance when evaluated by LIBERO.
The key technical breakthrough is the ability to generalize. Compared to DM0, zero-shot navigation success improved by 31%, multi-shot mode success improved by 45%, and fine-tuning success improved by 20%. The cost of fine-tuning decreased by 60%, allowing expert-level tasks to be performed on a single consumer RTX 4090 GPU in just 18 hours.
Three auxiliary platforms help bridge the engineering gap between model capabilities and real-world deployment. DFOL 2.0 utilizes the world model DW0.5 as a high-precision simulation for reinforcement learning in a virtual environment, reducing the need for real robot data by 60% and overall training costs by 40%. DexOS, the industry's first generalized operating system for embodied systems, decouples AI models from robot hardware, allowing developers to control, perceive, and connect robots with just a few lines of Python code, regardless of the robot brand.
The MaaS platform enables one-command zero-shot inference and provides a visual workspace for online training and deployment with pay-as-you-go billing.
The universal Apex robot platform, specifically designed for embodied AI applications, was also presented. It features a modular design with the ability to quickly swap end effectors in under 60 seconds, 360-degree scanning using two lidars, a 3 kg manipulator payload, millimeter precision, and hot-swappable batteries with a brain standby mode for continuous operation. The target MTBF exceeds 1000 hours.
CEO Tang Wenbin noted that embodied AI is at a critical inflection point where scaling laws fully demonstrate their power, but transitioning from models to real productivity requires overcoming the final hurdle of engineering integration. Dexmal is collaborating with robot manufacturers, including Tiangong, Huaqin, and Shihe, as well as chip partners such as Pingtouge, Moore Threads, and Cambricon, to build a domestic embodied AI ecosystem.
