ZTE Debuts 'AI FACTORY' Infrastructure

At MWC 2026, ZTE announced its 'AI FACTORY,' a co-designed AI infrastructure aimed at providing optimized computing power and efficiency for AI deployments. The solution is focused on delivering the best Total Cost of Ownership (TCO) for enterprises building out their AI capabilities.

ZTE's 'AI FACTORY' is built upon its full-stack intelligent computing solution, which encompasses everything from the underlying hardware to AI platforms and models. This includes high-performance training and inference servers, high-performance storage, and lossless networks, all designed to work in concert. The infrastructure is designed to be open and decoupled, allowing for compatibility with a wide range of mainstream GPUs and enabling on-demand resource allocation. A key component of this infrastructure is the "Elastic AIDC" (Artificial Intelligence Data Center) solution. This features a modular, prefabricated design that can reduce data center construction times by 40%. To tackle the high energy consumption of AI workloads, the solution utilizes advanced cooling technologies like cold-plate and immersion cooling, which can improve overall energy efficiency by 25%. The emphasis on Total Cost of Ownership (TCO) is addressed through a combination of hardware and software co-design. ZTE's AI Booster intelligent computing platform is designed to maximize the utilization of GPUs through features like automatic parallel training. This approach, combined with visual development and self-adapting parameter optimization, aims to lower the technical barriers for enterprises to train and deploy their own private AI models. The 'AI FACTORY' concept also extends to all-in-one solutions like the AiCube, which is a pre-integrated cabinet containing computing, storage, networking, and AI platform software. This out-of-the-box solution is aimed at simplifying the deployment of AI infrastructure, particularly for edge computing scenarios. It supports mainstream AI frameworks, allowing for more accessible training and inference of private AI models. At the software and application layer, ZTE is developing its Nebula Telecom Large Model and the Co-Sight AI Agent Factory. The Co-Sight platform provides a low-code development environment with modular components, enabling enterprises to rapidly create and deploy their own AI agents. This is part of ZTE's broader "AI for All" strategy, which aims to integrate AI across various industries. ZTE's own manufacturing facilities, such as its Binjiang factory, serve as a real-world testbed and showcase for these technologies. By implementing 5G, AI, and robotics in its own production lines, ZTE has achieved significant reductions in supply cycle times and energy consumption, demonstrating the potential benefits for other enterprises. This experience in smart manufacturing informs the development of their commercial AI solutions. The "co-designed" aspect of the 'AI FACTORY' appears to be rooted in ZTE's principle of "openness and decoupling," which involves collaboration with various partners. The company works with global partners to build a collaborative computing ecosystem. An example of this is the GTI 5G-A x AI Open Lab in Shanghai, where ZTE collaborates with partners like Qualcomm and Siemens to accelerate the integration of 5G and AI. This initiative is part of a wider presence at MWC 2026, where ZTE is showcasing a range of AI-powered innovations under the theme "Creating an Intelligent Future." These span from AI-native networks and AI-powered consumer devices to their full-stack cloud and computing solutions. The 'AI FACTORY' is a key pillar of their "Connectivity + Computing" strategy, aiming to provide the foundational infrastructure for the widespread adoption of AI.

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.