Share

Huawei Cloud reshaping industries with AI at GITEX GLOBAL 2023

Introducing Pangu Model 3.0, the first-of-its-kind commercial Arabic model
Huawei Cloud reshaping industries with AI at GITEX GLOBAL 2023
During GITEX GLOBAL 2023 (Supplied)

Huawei Cloud aims to create a substantial impact at GITEX GLOBAL 2023 by emphasizing ‘Everything as a Service’ and reshaping AI for various industries. Through the presentation of the ‘Pangu Model Reshaping AI for Industries’, Huawei Cloud showcases its dedication to constructing a cloud foundation that fosters an intelligent world through systematic innovation.

In anticipation of substantial growth in AI adoption across key industries, this strategic initiative aligns with PwC‘s Global Artificial Intelligence Study. The study projects that AI has the potential to contribute up to $15.7 trillion to the global economy by 2030.

Read more: Huawei to highlight win-win tech solutions at GITEX GLOBAL

During GITEX GLOBAL 2023, Huawei Cloud unveiled its Pangu Model 3.0, which comprises of the first-of-its-kind commercial Arabic model trained with hundreds of billions of parameters, surpassing peer models in understanding and generation capabilities.

Pangu model 3.0

Frank Dai, president of Huawei Cloud Middle East & Central Asia, said, “With our “Pangu model Reshaping AI for industries” solution, Huawei Cloud is showcasing our Pangu model 3.0 and Huawei Cloud Stack 8.3.0 at GITEX GLOBAL. The event offers us a unique opportunity to not only showcase our cutting-edge AI solutions but also to engage with industry leaders, innovators, and partners. It’s a dynamic environment where we can share our vision, exchange ideas, and gain valuable insights into the evolving needs of businesses and industries.”

Reshaping AI for Industries

Foundation models and related applications are the most complex hardware and software system engineering, which requires systematic innovation. Roy Luo, vice president of Consulting and Solution of Huawei Cloud Middle East & Central Asia, said: “Huawei Cloud is committed to reshaping AI for Industries and bringing the most advanced cloud and AI technologies to the Middle East and Central Asia. Huawei Cloud focuses on systematic innovation. At the system architecture layer, the distributed QingTian architecture breaks the boundaries of computing, storage, and network, the traditional primary/secondary architecture is upgraded to the peer-to-peer architecture, greatly improving AI computing power. At the data processing layer, Huawei Cloud provides one-stop technologies for the entire cloud process, including data storage, analysis, sharing, and labeling, enabling enterprises to find, store, remember, and use foundation model data. We are committed to empowering enterprises from different industries in the region to build their own large models”.

Integral part

Ushering into the AI era, AI is transforming how we live, work, and interact. Dr. Zhu Shenggao, vice president of AI Business at Huawei Cloud Middle East & Central Asia, stated: “AI has become an integral part of our society, permeating every aspect of our lives. Its adoption in industries is undergoing a rapid transformation. This remarkable shift is being driven by the emergence of large AI models meticulously trained on massive datasets. It is not just a technological advancement, but a fundamental shift that will shape the future of industries, and Huawei Cloud is at the forefront of enabling this transformation.

Meticulous design

Huawei Cloud’s Pangu models are meticulously designed to address the practical requirements of specific industry scenarios. The latest release, Pangu Models 3.0, introduces a 5+N+X three-layer decoupled hierarchical architecture. This architecture comprises five L0 foundation models, N industry-tailored models at the L1 layer, and X AI applications at the L2 layer. Together, they establish a robust foundation for a wide array of models and applications. The L0 layer of foundation models encompasses NLP, CV, multimodal, prediction, and scientific computing, providing versatile capabilities to fuel a multitude of industry-specific applications.

50 billion tokens

With over 1,000 AI projects, Huawei Cloud is steadfast in its commitment to the “AI for Industries” strategy. Pangu models have been meticulously pre-trained using open datasets from over 10 industries, such as Finance, Government, Meteorology, Healthcare, Education, Automotive, and Utilities. The data volume for each industry surpasses 50 billion tokens. On October 18, Dr. Zhu will be a part of the AI Everything keynote, delving into the acceleration of AI for industries through Pangu Large Models.

Huawei Cloud Stack 8.3.0

During GITEX GLOBAL, Huawei Cloud will showcase its latest offering, Huawei Cloud Stack 8.3.0. This hybrid cloud platform is designed to support the convergence of data and AI, enabling seamless deployment of AI cloud services within on-premises data centers. With over 100 cloud services, Huawei Cloud Stack 8.3.0 augments the core capabilities of the Pangu Large Model, empowering customers to explore the full potential of cloud computing. Renowned as the most comprehensive AI production chain in the industry, it significantly reduces the entry barrier for large-scale model development.

Flexibility and autonomy

Deploying Pangu Model 3.0 offers customers the flexibility and autonomy to create models on both the public cloud and on-premises infrastructure. With the added benefits of Huawei Cloud Stack 8.3.0, enterprises gain access to a comprehensive AI pipeline that streamlines the efficient development of large-scale models.

Huawei Cloud’s GaussDB

In addition, Huawei Cloud’s GaussDB distributed database product stands out for its exceptional core capabilities in serverless, region-less, and mode-less functions. This enables the provision of data services that exhibit unparalleled availability, performance, scalability, security, and intelligence.

For more news on telecom, click here.

The stories on our website are intended for informational purposes only. Those with finance, investment, tax or legal content are not to be taken as financial advice or recommendation. Refer to our full disclaimer policy here.