Huawei, a leading technology company, is paving the way for new opportunities and technological advancements across various sectors with the introduction of the highly anticipated Pangu Model 3.0 and Ascend AI cloud services. Mr. Zhang Pingan, Huawei Executive Director and CEO of HUAWEI CLOUD, unveiled the new AI model during his keynote speech at Huawei’s Developer Conference, reaffirming Huawei’s commitment to empowering industry customers and partners while harnessing the transformative potential of artificial intelligence.
Pangu 3.0 is a system of pre-trained models that can be quickly adapted to meet scenario-specific needs and address complex challenges across multiple industries. By leveraging large-scale data sets and machine learning algorithms, Pangu 3.0 is set to revolutionize the industrial application of AI in diverse areas, such as weather forecasting, drug development, fault identification in trains, and the mining industry.
The Pangu model 3.0 consists of three capability segments: L0, L1, and L2. The L0 capability segment represents the general capability of the large model. It possesses exceptional learning abilities, allowing it to quickly acquire new skills such as weather forecasting and drug development with minimal training. The L1 capability segment caters to industry-specific applications, while the L2 capability segment focuses on business scenarios, such as financial exception detection, file review, and wave forecasting. In simpler terms, L0 is a generalist, L1 is a professional, and L2 is a specialist. Enterprises can choose the specific capability segment that aligns with their requirements. The general model represents the human brain, while the industrial model serves as the hands and feet, excelling in concrete tasks. It is evident that the future of large models lies in solving complex problems and carrying out specific tasks.
For instance, quality inspection on an industrial assembly line may seem inconsequential, but the collective experience of numerous enterprises and assembly lines can lead to the development of robust solutions. The deployment of large models, therefore, represents the beginning of a new era in training.
Mr. Zhang stated, “Huawei Cloud Pangu models will empower individuals across every industry, providing them with an intelligent assistant that enhances productivity and efficiency. Upholding our mission of ‘AI for Industries,’ we will utilize Pangu models to reshape various sectors through AI.”
The prestigious scientific journal Nature recently published a paper highlighting the vast potential of artificial intelligence in weather prediction. Huawei’s Pangu Meteorology Model team developed a highly precise and accurate global AI weather forecast system based on deep learning, utilizing 43 years of data. This system can run on a basic desktop computer and outperforms traditional numerical forecast models by 20% in terms of accuracy while delivering results 10,000 times faster. The Pangu Meteorology Model focuses on key elements and common time ranges, enabling more precise weather forecasts including humidity, wind speed, temperature, and sea level pressure. Notably, it accurately predicted the trajectories and landing times of typhoons, including Typhoon Mawar in May of this year, showcasing its exceptional capabilities and wide range of applications.
Dr. Tian Qi, Chief Scientist of HUAWEI CLOUD AI Field, an Academician of the International Eurasian Academy of Sciences, “Weather forecasting is one of the most challenging scenarios in scientific computing due to its complex nature and the vast mathematical and physical knowledge it requires. AI models can extract statistical laws of atmospheric evolution from massive data. Currently, Pangu-Weather focuses on enhancing the forecasting system by predicting the evolution of atmospheric states. Our ultimate goal is to build next-generation weather forecasting frameworks using AI technologies to strengthen existing systems.”
The railway industry also benefits significantly from Huawei’s AI advancements and the implementation of the Pangu Model 3.0. The Pangu Railway Model greatly improves the safety and efficiency of freight trains. Previously, fault identification within Train Freight Detection Systems (TFDS) was labor-intensive, inefficient, and costly. However, with the introduction of the Pangu Railway Model, the process has been revolutionized, resulting in the accurate identification of 442 types of faults. The model achieves detection rates surpassing 99.99% for major faults and exceeding 98% for common faults.
Furthermore, the Pangu Model 3.0 finds valuable applications in the mining industry. Traditional mining poses significant risks, requiring extensive manpower and facing technical and safety challenges. However, by leveraging digital technologies and the Pangu Mining Model, the industry is experiencing a transformative shift. With massive amounts of data for model pre-training, the Pangu Mining Model enables unsupervised self-learning, covering over 1,000 sub-scenarios in coal mining operations. From digging and drivage to machinery, transport, and communication, the model streamlines AI applications across numerous coal mining scenarios.
By harnessing the capabilities of Huawei’s Pangu 3.0 and Ascend AI cloud services, businesses can unlock new levels of safety, efficiency, and productivity, driving substantial growth and gaining a competitive advantage in their respective industries. Huawei’s commitment to innovation and the development of advanced AI solutions underscores its dedication to providing cutting-edge solutions for its customers.