According to industry media reports, with the increase in artificial intelligence research and development and application, demand for high-performance GPUs for data centers, such as the Nvidia H100 and A100, has also increased dramatically. Many companies have made large purchases, and Nvidia's high-performance GPUs are also in short supply. Indian data center and server operator Yotta, which ordered 16,000 H100 units from Nvidia last year, will order another 16,000 GPUs from Nvidia. Yotta's new GPU purchase order will continue until March next year. In addition to the H100, there will also be an H200. The order amount is expected to be around 500 million US dollars, which is comparable to last year's order amount.

Under the wave of big AI models, large models and vertical applications are driving the demand for AI computing power to continue to increase. Cinda Securities believes that with the gradual development of AI, demand for downstream computing power continues to rise, the closed loop on the application side is gradually becoming clear, and it is firmly optimistic about the computing power industry chain.

According to the Finance Federation's theme library, among the relevant listed companies:

IFF's AI servers have been iterated to the fourth generation, and the development and mass production of high-performance AI servers such as Nvidia's H100 and H800 for customers began this year.

The “Zhaohan” series of AI servers produced by Topway Information can meet the training needs of various mainstream AI scenarios and AI models, and have been used in many regional artificial intelligence computing centers, urban artificial intelligence centers, and general AI server scenarios in China.