Industry trends
2026.07.15
On July 9th, Luculent showcased its latest AI achievements at the Photosynthesis Organization 2026 Intelligent Computing Application Conference. With the theme of "Boundless Intelligent Computing, Paradigm Shift", the conference brought together leading domestic computing power enterprises, industrial digital service providers, scientific research institutions and end-users from various industries. Centering on the new development paradigm of intelligent computing, participants conducted in-depth discussions on the integrated implementation of computing power infrastructure, artificial intelligence algorithms and physical industry scenarios, and jointly explored new paths for collaborative innovation of domestic software and hardware, construction of green computing power infrastructure and large-scale application of industrial AI.
During the conference, Luculent and Hygon Information signed a strategic cooperation agreement. Relying on Hygon Information's R&D and industrialization advantages in the field of domestic CPU and DCU core computing hardware, and combining Luculent's software and scenario accumulation in the industrial digital and intelligent field, the two parties will closely follow the national development strategy of independent and controllable software and hardware, integrate core technologies and industry resources of the upstream and downstream of the industrial chain, jointly tackle the localization replacement of key core technologies, and consolidate the foundation of domestic industrial information security.
Based on the strategic cooperation consensus, the two parties will jointly build the "Hygon-Luculent Xinchuang Joint Adaptation Laboratory", complete the in-depth adaptation and optimization of Hygon Information's CPU and DCU core computing power hardware with Luculent's full range of industrial software and industry solutions, and create a benchmark solution for integrated domestic software and hardware with stability, reliability and leading performance. Joint R&D will be carried out in cutting-edge directions such as green computing, next-generation data center architecture, edge intelligence and industrial large models to accelerate the large-scale implementation of green and efficient domestic computing power in industrial scenarios.
Subsequently, the two parties jointly released the Hygon·Luculent Sunshine Yaoguang Industrial Large Model All-in-One Machine.
Based on the needs of private and secure deployment in industrial enterprises, this all-in-one machine has four core advantages: professional model, strong computing power, safety and reliability, and fast implementation. It directly addresses the pain points of digital transformation in industrial enterprises with a closed-loop innovation model of "data + algorithm + scenario". Relying on the full-stack domestic hardware base and mature industrial vertical AI capabilities, it provides factories with a lightweight, implementable and high-value one-stop industrial intelligent computing carrier, which greatly reduces the technical threshold and implementation cycle for industrial enterprises to build large model applications.
At the conference, experts from Luculent delivered a keynote speech titled "Empowering 'Small Scenarios' with AI, Building a Smart 'Big Landscape' in the Energy Industry". The speech pointed out that industrial digital and intelligent transformation has become an inevitable path for industry development at the current stage, but the industry generally faces multiple obstacles such as weak industrial data foundation, scarce mature implementation models, high difficulty in cross-scenario model reuse, and insufficient supply of compound technical talents, which restrict the popularization of industrial AI implementation.
Luculent proposed a three-in-one implementation idea of platformization, scenarization and valuation. Relying on the one-stop AI platform and scenario-based application services, the company focuses on high-value business fields such as AI+equipment, AI+energy consumption, AI+safety, AI+quality and AI+operation, and gradually builds an industry intelligent ecosystem covering the entire industrial chain.
In the speech, the experts explained the value of scenario-based AI combined with real enterprise implementation cases. In the field of intelligent operation and maintenance of key equipment, relying on the energy vertical industrial large model and multi-agent collaboration, Luculent has built a full-process intelligent equipment diagnosis and operation and maintenance system for enterprises such as China Coal Equipment, State Nuclear Demonstration and Ningbo Iron and Steel, realizing equipment fault early warning, root cause diagnosis and intelligent generation of maintenance solutions. In the scenario of energy-saving optimization for high-energy-consuming equipment, an energy consumption optimization intelligent agent has been created by integrating reinforcement learning and visual recognition algorithms, realizing dynamic and precise regulation of dust removal fans. This solution has been implemented in more than 20 steel enterprises including Meigang and Kungang, with an average annual power saving of 450,000 kWh and carbon emission reduction of 260 tons per fan, and the energy-saving range can reach 10%-15%. The relevant technical achievements have been included in the "List of Key Projects" of the Ministry of Industry and Information Technology and the Extreme Energy Efficiency Capability List of the China Iron and Steel Association.
The cooperation and achievement release at this conference mark another key step in Luculent's layout of domestic industrial AI. In the future, the company will continue to join hands with leading domestic computing power enterprises such as Hygon, continuously improve the independent and controllable industrial large model software and hardware product system, deeply cultivate the segmented scenarios of the process industry, and drive the high-quality and green development of China's industrial industry with domestic intelligent computing power.
