event:

in the early morning of February 22, Beijing time, Avida disclosed its financial results for the fourth quarter and full year of fiscal year 2024 and released its revenue outlook for the first quarter of 2025.

review:

data center-driven single-quarter revenue continued to hit a record high, profitability exceeded the guidance cap. According to NVIDIA's financial report, the company's revenue in the fourth quarter of 2024 was US $22.1 billion, an increase of 265 percent over the same period last year and 22 percent over the previous quarter, exceeding market expectations of US $20.41 billion, a record high; the company's single-quarter net profit was US $12.285 billion billion, an increase of 769 percent over the same period last year; the gross profit margin in a single quarter was 76.7 percent, up 10.6 percent over the same period last year and 1.7 percent, exceeded the upper limit of the previously expected guideline (75.5 per cent fluctuated by 0.5 percentage points). In terms of business, the growth of data center business is the main driving force for the company's substantial revenue growth. The company's data center revenue in a single quarter was 18.4 billion billion US dollars, up 409 percent year-on-year and 27% month-on-month, setting a new record. According to the company's disclosure, large cloud service providers account for more than half of the revenue in the data center business, benefiting from the continued popularity of generative artificial intelligence (AI) models. The demand for AI training, reasoning and data processing from large cloud service providers, GPU providers, enterprise software and consumer Internet companies exceeds market expectations. Hopper GPU platform drives the data center business to achieve strong growth, driving the company's single-quarter revenue to a record high.

AI computing power demand continues to be hot, Avida FY25Q1 outlook optimistic. As a global GPU (graphics processing unit) leader, Nvidia provides high-performance data centers, cloud services, PC workstations, automotive self-driving vehicles and embedded system support for a variety of devices through its deep learning and AI solutions. According to Jon Peddie Research(JPR) data, the company's global independent GPU market share in the first quarter of 2023 was 87%, significantly ahead of its competitors AMD(10%) and Intel (3%), leading the industry. Huang Renxun, founder and CEO of Avida, said in mid -2023 that the next wave of AI will be artificial intelligence, that is, intelligent systems that can understand, reason and interact with the physical world, and will put forward higher requirements for computing power. At the end of 2023, the company released a new generation of GPU H200, whose theoretical performance reached twice that of H100, and said it was expected to ship in FY25Q2. The company said that thanks to the high demand for its industry-leading artificial intelligence chips and the expected improvement in supply chain dynamics, artificial intelligence-related businesses are expected to continue to be in short supply in 2024. It looks forward to revenue of US $24 billion in the first fiscal quarter of 2025 (fluctuating by about 2%), up about 9% month on month, exceeding the market expectation of US $22.1 billion. Gross profit margin is about 77% (up and down 0.5), up slightly.

Nvidia is a global AI industry bellwether, focusing on server hardware investment opportunities and independent computing power-related companies. As a bellwether of the global artificial intelligence industry, NVIDIA FY24Q1 revenue performance hit a record high, highlighting its strong position in the AI field and the continued boom in downstream demand. As the underlying cornerstone of the AI era, computing power is expected to once again arouse market attention, the relevant suppliers are expected to usher in the release of performance. It is recommended to pay attention to the data center hardware level of AI chips, server PCB, IC carrier board, DDR and advanced packaging and other sub-benefit areas, in addition to the domestic CPU/GPU and other independent computing industry chain related enterprises. Such as Haiguang Information (688041), Cambrian (688256), Shenghong Technology (300476), Shanghai Electric (002463), Xingsen Technology (002436), Lanqi Technology (688008) and Tongfu Micropower (002156).

Risk Warning: Downstream commercialization landing speed is not as fast as expected, upstream supply is not as fast as expected, etc.