On September 22-23, the Intel Connection 2026 Conference on Technology Innovation and Industrial Ecology will be held at the Suzhou International Expo Center. Lixun Precision unveiled its AI Infra series product NEXTiX LEGION at the conference, showcasing a variety of AI computing power machines, as well as comprehensive solutions covering core components, prefabricated data centers, and computing power services. Relying on its advantages in technology research and development, precision manufacturing, and vertical integration, Luxshare Precision continues to deepen its AI computing infrastructure layout and continuously improve its standardized and scalable delivery capabilities.

With the continuous expansion of large-scale models and the continuous enrichment of application scenarios, computing infrastructure is evolving from standalone to systematic and clustered. Around this trend, NEXTiX LEGION adheres to the product concept of "changing with the model, fighting for the scene", and is built on a unified iX Architecture architecture. It focuses on four major capabilities: intelligent scheduling (Intelligence), efficient computing power (eXecution), high-speed interconnection (eXchange), and extreme expansion (eXtreme), achieving flexible expansion from a single server to a 10000 card GPU cluster, providing efficient and scalable computing power support for diverse AI application scenarios.

Legion based complete machine series, covering diverse computing power scenarios



9月22日至23日，英特尔技术创新与产业生态大会（Intel Connection 2026）在苏州国际博览中心举行。立讯精密携 AI Infra 系列产品 NEXTiX LEGION 亮相大会，集中展示多款 AI 算力整机，以及覆盖核心部件、预制数据中心和算力服务的整体解决方案。依托技术研发、精密制造与垂直整合优势，立讯精密持续深化 AI 算力基础设施布局，并不断提升标准化、规模化交付能力。

随着大模型规模不断扩大、应用场景持续丰富，算力基础设施正由单机向系统化、集群化方向演进。围绕这一趋势，NEXTiX LEGION 以“随模型而变，为场景而战”为产品理念，基于统一的 iX Architecture 架构打造，围绕智能调度（Intelligence）、高效算力（eXecution）、高速互联（eXchange）与极限扩展（eXtreme）四大能力，实现从单台服务器到万卡级 GPU 集群的灵活扩展，为多元 AI 应用场景提供高效、可扩展的算力支撑。

军团化整机系列，覆盖多元算力场景



At the conference, six NEXTiX LEGION products, including TITAN, APEX, NOVA, IRON, VORTEX, and FALCON, were showcased, covering various application scenarios such as large-scale computing centers, core training, open heterogeneous, enterprise level AI, high-density computing, and high-density liquid cooling.

The series of products are built on a unified architecture, maintaining collaboration in firmware, heat dissipation design, operation and maintenance system, and cluster networking, which can effectively reduce the complexity of large-scale deployment and later operation and maintenance, providing more flexible and efficient computing power choices for different scales and application needs.

Prefabricated data centers accelerate the deployment of computing power



本次大会现场，TITAN、APEX、NOVA、IRON、VORTEX、FALCON 六款 NEXTiX LEGION 产品集中亮相，覆盖大型算力中心、核心训练、开放异构、企业级 AI、高密计算及高密液冷等多元应用场景。

系列产品基于统一架构打造，在固件、散热设计、运维体系及集群组网等方面保持协同，可有效降低大规模部署与后期运维复杂度，为不同规模、不同应用需求提供更加灵活、高效的算力选择。

预制化数据中心，加速算力部署



Facing the rapidly growing demand for computing power, Lite On Precision continues to promote a prefabricated and modular data center construction model. By integrating and debugging power supply, heat dissipation, computing power, networking, and storage systems in advance, and then transporting them to the project site for modular deployment, it effectively shortens the data center construction cycle and can flexibly expand various functional modules according to business needs, enabling new resources to be quickly converted into available computing power and further improving the efficiency and flexibility of infrastructure construction.

Extending from product delivery to computing power services

While continuously improving hardware products and infrastructure solutions, Luxshare Precision actively expands its computing power service model. By building a GPU computing power resource pool and combining remote monitoring, on-site operation and maintenance capabilities, it flexibly configures computing power resources according to actual business needs, promoting the extension of computing power from product delivery to continuous service. This exhibition further showcases Luxshare Precision's layout in the fields of AI computing power machines, liquid cooling infrastructure, prefabricated data centers, and computing power services, reflecting the group's comprehensive capabilities that continue to extend from core components to machines, systems, and services.



面向快速增长的算力需求，立讯精密持续推进预制化、模块化的数据中心建设模式，通过将供电、散热、算力、组网及存储等系统提前完成集成与调试，再运输至项目现场进行模块化部署，有效缩短数据中心建设周期，并可根据业务需求灵活扩展各功能模块，使新增资源更快转化为可用算力，进一步提升基础设施建设的效率与灵活性。

从产品交付向算力服务延伸

在持续完善硬件产品与基础设施解决方案的同时，立讯精密积极拓展算力服务模式，通过构建 GPU 算力资源池，并结合远程监控、现场运维等服务能力，根据实际业务需求灵活配置算力资源，推动算力能力由产品交付向持续服务延伸。此次参展进一步呈现了立讯精密在 AI 算力整机、液冷基础设施、预制数据中心及算力服务等领域的布局，体现集团从核心部件向整机、系统及服务不断延伸的综合能力。



In the future, Luxshare Precision will continue to leverage its comprehensive advantages in technology research and development, precision manufacturing, vertical integration, and large-scale delivery, and work with industry chain partners to continuously improve AI computing infrastructure products and solutions, providing reliable and efficient computing support for the intelligent transformation of thousands of industries.



未来，立讯精密将继续发挥在技术研发、精密制造、垂直整合及规模化交付等方面的综合优势，携手产业链合作伙伴，持续完善 AI 算力基础设施产品与解决方案，为千行百业智能化转型提供可靠、高效的算力支撑。

(Using AI translation)

（使用AI翻译）