Huawei Cloud: Fostering the Fertile Ground for Compute, Empowering AI Pioneers for Industries
Sep 19, 2025
[Shanghai, China, September 19, 2025] On the second day of HUAWEI CONNECT 2025, Zhang Ping'an, Huawei's Executive Director of the Board and CEO of Huawei Cloud, delivered a keynote speech titled "All Intelligence: Empowering AI Pioneers for Industries". He shared Huawei Cloud's innovation and practices in AI compute services, foundation models, embodied AI, AI agents, and much more. Additionally, Charles Yang, Senior Vice President of Huawei and President of Huawei Cloud Global Marketing and Sales Service, shared Huawei Cloud's innovations and experiences that are empowering enterprises to succeed in global markets in the digital and intelligent era.
Constant innovation in AI Compute Service: Unleashing powerful compute in the intelligent era
This year, Huawei Cloud announced its AI Compute Service powered by CloudMatrix384. The specifications of the Huawei CloudMatrix supernode will be upgraded from 384 cards to 8,192 cards. The supernodes can support a hyperscale cluster running on 500,000 to 1 million cards, thus providing robust AI compute, an invaluable resource in the intelligent era. Huawei Cloud also announced innovative memory storage with its Elastic Memory Service (EMS), which achieved an industry first by expanding video RAM with memory. This drastically reduces the latency of multi-round conversations on foundation models, greatly improving user experience.
With rapidly increasing demand for AI compute, traditional data centers are no longer sufficient to serve as efficient AI infrastructure. To support AI workloads, the power consumption of a single cabinet will need to be upgraded from 10 kW to 70 kW, and possibly as high as 200 kW, and heat dissipation will need to shift from air cooling to full liquid cooling. Huawei Cloud has deployed fully liquid-cooled AI data centers in China's Guizhou, Inner Mongolia, and Anhui. These AI data centers support 80 kW heat dissipation per cabinet, reduce power usage effectiveness (PUE) to 1.1, and offer AI-enabled O&M. This means enterprises do not need to reconstruct traditional data centers or build new ones. Instead, they require only a pair of optical fibers in order to connect to the data center and access efficient AI compute, as well as full-stack dedicated AI cloud services, on Huawei Cloud.
Zhang Ping'an pointed out that Huawei Cloud's AI Token Service abstracts away the underlying technical complexity and directly provides users with the final AI computing results. This allows users to utilize the inference computing power in the most efficient way possible. The CloudMatrix384 supernode realizes the full pooling of compute, memory, and storage resources, decouples compute tasks, storage tasks, and AI expert systems, and converts serial tasks into distributed parallel tasks, greatly improving the inference performance of the system. In scenarios involving inference tasks with different latency requirements, such as online, nearline, and offline inference, CloudMatrix384 delivers an average inference performance per card that is 3 to 4 times that of H20.
At the conference, Zhang Ping'an announced the official launch of the AI Token Service powered by CloudMatrix384. The service delivers superior performance, service, and quality to customers.
As of now, the number of customers worldwide using Huawei Cloud's AI Compute Service has increased from 321 in 2024 to this year's 1,805. Qihoo 360's Nano Search has used the CloudMatrix384-powered token inference service to process tens of thousands of content generation requests every day.
Zhang Ping'an noted that Huawei Cloud will continue to advance software-hardware synergy and architecture innovation, so as to build cloud services that constantly create value for their customers and always keep them up-to-date with the latest technologies. In addition, Huawei Cloud will invest more resources into AI and computing, in order to cultivate the fertile ground on which different industries' applications can run smoother and better. In the AI era, Huawei Cloud is committed to pursuing innovation in AI computing power and technologies, helping every customer become an AI pioneer in their industry.
Zhang Ping'an, Huawei's Executive Director of the Board and CEO of Huawei Cloud
Tackling challenges head-on: Helping enterprises build their own models
Huawei Cloud has been honing its Pangu Models by diving into industry-specific scenarios, and has worked with its customers to tackle their most pressing challenges head-on, reimagining what is possible in these industries. Huawei uses openPangu to provide best practices for AI training and inference, making it easier for developers to efficiently use AI computing power. Zhang Ping'an noted that, at the same time, Huawei is developing the closed-source Pangu Model. Huawei will continually increase investment in Pangu Models, constantly study industry scenarios to better understand customer requirements, and support customers in developing their own industry-specific models, thus accelerating intelligent transformation across industries.
To better align the incremental training of models with enterprise scenarios, Huawei Cloud integrates industry know-how into ModelArts, a one-stop AI development platform. ModelArts provides knowledge regarding the optimal data ratios for industry tasks, the optimal configurations for incremental training, and a system for assessing model effectiveness and accuracy.
Zhang Ping'an also pointed out that data quality determines model quality. Huawei Cloud uses the DataArts data governance pipeline to automatically extract multi-modal data semantics and build converged knowledge graphs that form a unified knowledge lake for enterprises. This provides comprehensive, rich, and consistent business corpuses for large model pre-training. In addition, Huawei Cloud provides one-stop toolchains to quickly generate high-quality tuning datasets with chains of thought (CoT) for model post-training.
China Southern Airlines uses the Pangu Model's cross-modal prediction technology to predict passenger traffic and average ticket prices over an 18-month period. Prediction accuracy reaches 90%, significantly improving the airline's operating revenue. The Chinese Academy of Agricultural Sciences (CAAS) has developed the Agricultural Scientific Discovery Model based on the Pangu foundation model and Pangu Scientific Computing Model. This model is trained using massive amounts of agricultural knowledge and tens of billions of biological data entries, shortening the R&D cycle by more than 50%.
Pangu Models have been applied in more than 500 scenarios across over 30 industries. They have played a significant role in fields like government services, finance, manufacturing, healthcare, coal mining, steel, railways, autonomous driving, and meteorology.
Moving beyond terminals: Enabling infinite intelligence evolution on the cloud
This year, Huawei Cloud launched the CloudRobo Embodied AI Platform, which deploys complex algorithms and intelligent logic on the cloud to realize more lightweight robots. By taking advantage of the massive computing power and advanced AI models on the cloud, the platform makes robot execution more intelligent. Cloud intelligence overcomes the limitations that have been holding robots back, making them applicable to more scenarios.
In the development phase, CloudRobo uses a cloud-based, embodied AI training data generation platform to automatically generate operation trajectories for different poses, textures, and lighting conditions, greatly enriching data diversity and increasing the synthetic data scale by thousands of times. This not only greatly reduces data collection costs, but effectively improves the generalization capability of models. CloudRobo also provides cloud-based environment perception, global planning, and autonomous execution capabilities, achieving a task execution accuracy exceeding 90%. With the support of CloudRobo, the National and Local Co-Built Humanoid Robotics Innovation Center in Shanghai uses data collected from real machines alongside synthetic data to train robots, improving the overall sorting success rates of robots in multiple scenarios to over 90%, which is an industry-leading success rate so far.
To build a unified, open, and secure communication channel between robots and the cloud, Huawei Cloud has launched the Robot to Cloud (R2C) Protocol, which realizes the collaborative thinking and efficient execution of AI agents. As of today, national standards are in progress regarding data collection, data generation, and communication interfaces around the R2C Protocol. During his keynote speech, Zhang Ping'an announced that the first 20 partners of the R2C Protocol were officially onboard, including Hualong, Flexiv, Yijiahe, and Avant Robotics. Huawei Cloud works alongside its partners to develop cloud-based integrated robot solutions based on the R2C Protocol in fields such as manufacturing, logistics, inspection, and healthcare.
Kunpeng Cloud Services: Empowering industry innovation with software-hardware synergy and an open ecosystem
One of Huawei Cloud's key strategies is to develop Kunpeng-powered ARM cloud services that deliver performance, security, and reliability. In the past year, the number of Kunpeng compute cores on Huawei Cloud has increased from 9 million to 15 million, an increase of 67%. In addition, the Kunpeng platform has been continuously improved to support compatibility with mainstream software, and has also been adapted to more than 25,000 applications. The Kunpeng platform provides solid support for Kunpeng Cloud Services to be applied to even more general-computing scenarios, in addition to transcoding, databases, web applications, and cloud phones.
GaussDB: Building efficient, reliable data foundations based on supernodes and full pooling
Based on general-purpose computing supernodes, Huawei Cloud's GaussDB databases realize the layered pooling of compute, memory, and storage resources, and allow multi-read and multi-write on any node at the same time, breaking free from the restrictions of the traditional architecture where only the primary node supports data read/write. GaussDB databases also support dynamic load scheduling, greatly enhancing the performance of concurrent transaction processing. A GaussDB cluster deployed based on computing supernodes can process 5.4 million transactions per minute, marking a 2.9-fold performance increase over non-supernode clusters.
All-scenario distributed cloud: Ubiquitous and best possible compute with local access
Huawei Cloud has built a distributed cloud solution covering all scenarios, including CloudOcean, CloudSea, CloudLake, and CloudPond. This covers central regions, hotspot areas, and edge sites, bringing consistent Huawei Cloud experience to wherever customers' business is located. Huawei Cloud and FAW-Volkswagen have worked together to successfully implement China's first automotive factory based on the distributed cloud solution. The solution overcomes geographical constraints and realizes comprehensive, in-depth collaboration between the HQ and six factories spread across five cities, enabling seamless data flow and intelligent manufacturing.
Building an easy-to-use, effective, and open platform for developing and running agents
AI agents are transforming into a new application paradigm in the AI era. With this in mind, Huawei Cloud has launched Versatile, an enterprise-grade agent platform. It aims to serve as an easy-to-use, effective, and open platform for developing and running AI agents. With this platform, customers will be better equipped to quickly develop AI agents that suit their application scenarios.
Many nodes are included in the agent development process, involving highly complex businesses. Based on Versatile, users simply need to prepare and enter business description documents and flowcharts. After simple confirmation, the agent can be generated in two steps, greatly enhancing generation efficiency. Agents automatically invoke numerous programs when they are working, causing severe traffic jitters. Versatile also uses system innovations such as high-performance sandbox and full-memory cache to reduce end-to-end response latency within businesses by more than 40%, enabling agents to run stably and efficiently.
Continuous innovation: Empowering customers to succeed in global markets
As industries worldwide rush to embrace digital and intelligent transformation, Huawei Cloud consistently pursues technological innovation as its core driving force and is committed to providing stable, efficient, and secure cloud services for global customers.
Charles Yang, Senior Vice President of Huawei and President of Huawei Cloud Global Marketing and Sales Service, delivered a keynote speech regarding this topic. He highlighted the fact that enterprises are facing numerous complex challenges in the digital intelligence era. With this in mind, Huawei Cloud identifies its core positioning and adheres to this positioning. Huawei Cloud provides a holistic support system featuring "one global cloud infrastructure and two engines – one with data and one with AI" for its global customers.
The first of these is KooVerse, a global cloud infrastructure that offers customers elastic cloud services. Second are data enablement services, which efficiently supply data and unleash data value. Third is a one-stop AI development platform that helps enterprises build AI-native applications and innovate with greater ease. Finally, Huawei Cloud brings together the best practices in global digital and intelligent transformation, and provides them through cloud services to help more customers succeed.
Charles Yang, Senior Vice President of Huawei and President of Huawei Cloud Global Marketing and Sales Service
DeFacto, a top fast fashion retailer in Türkiye, faces a traffic surge of more than 10 times during each promotional season, and its traditional IDC architecture struggled to cope with this surge. DeFacto therefore turned to Huawei Cloud's container service, which can automatically scale up 4,000 pods within 30 seconds, allowing the company to easily cope with traffic spikes.
Based on Huawei Cloud's DWS, Neogrid, Brazil's largest data intelligence and technology company, migrated and integrated its data that was scattered across multiple platforms, improving data integration efficiency by 40% and data analysis efficiency by 50%. Today, this gives Neogrid's customers the ability to obtain the day's data before the end of work and therefore make business decisions earlier.
Based on the powerful generalization capability of Huawei Cloud's Pangu Prediction Model on industrial time-series data, Wanhua Chemical, a global leader in the chemical industry, has accurately captured data on more than 2,000 key devices, allowing them to transform their reactive maintenance model into predictive maintenance. Model prediction accuracy has improved from 70% to 90%, and anomaly identification efficiency has increased by 10%.
Huawei worked with partners such as China Unicom to build "One Cloud, One Network, One Platform" for Changan Automobile, driving flexible manufacturing with data. Changan Automobile uses 5G and IoT to connect 12,000 devices, allowing it to interconnect data from different domains. Based on the comprehensive data services of Huawei Cloud, Changan has built a unified data platform. This platform breaks apart the data silos, enables efficient production operations, and shortens order delivery lead time from 21 days to 15 days.
In addition to providing end-to-end service capabilities that span the cloud foundation, technology enablement toolchains, and experiences, Huawei Cloud provides tailored consulting and professional services, as well as automation tools and platforms for enterprises' digital and intelligent transformation.
Charles Yang expressed that digital and intelligent transformation can never be achieved by one player in a single move. Rather, it requires close, continuous collaboration throughout the entire journey. With its technologies, experiences, and services, Huawei Cloud is ready to work with global enterprises at every step of their transformation journeys.
HUAWEI CONNECT 2025 runs from September 18 to 20 at the Shanghai World Expo Exhibition & Convention Center and Shanghai Expo Center. This year, the event explores the theme of "All Intelligence" and delved into AI across three dimensions: strategy, technology, and ecosystems. This involved an in-depth look at Huawei's latest strategic initiatives, and the unveiling of a series of all-new digital and intelligent infrastructure products, scenario-specific solutions for industries, and development tools.
In addition to keynote speeches given at the event, Huawei Cloud brought a variety of agendas, such as summit forums and roundtables. Huawei Cloud also works with customers and partners to exhibit a wide array of innovative technologies and practices in fields such as cloud infrastructure, large models, databases, AI agents, and embodied AI, demonstrating how technology can facilitate the digital and intelligent transformation of industries.