Cambrian's performance soars! Zhang Jianping continues to add to his holdings

【Related Reading】Cambricon’s revenue grew over 400% last year, with “market influencer” Zhang Jianping becoming the fifth largest shareholder

Over 2 billion! Cambricon reports its first annual profit! Plans to pay a 15 RMB dividend per 10 shares and transfer 4.9 shares for every 10 shares

First full-year profit! Cambricon’s 2025 revenue soars 453%, with a surge in inventory and customer concentration raising concerns

On the evening of March 12, Cambricon released its 2025 annual report. The announcement shows the company achieved operating revenue of 6.497 billion RMB, a year-on-year increase of 453.21%; net profit attributable to shareholders of the listed company was 2.059 billion RMB, marking its first full-year profit and turning losses into gains.

Notably, according to Cambricon’s annual report, Zhang Jianping increased his holdings to 6.8149 million shares, accounting for 1.62% of the circulating shares. Based on Cambricon’s closing price of 1,099 RMB today, Zhang Jianping’s holdings are worth nearly 7.5 billion RMB.

Multiple Industry Deployments at Scale

Products Gaining Wide Recognition

The annual report indicates that during the reporting period, relying on significant progress in AI chip products, basic software platforms, and cluster software toolchains, the company’s products have been deployed at scale in key industries such as telecommunications, finance, and internet. They have passed rigorous customer environment tests, and their versatility, stability, and ease of use have been widely recognized by clients.

Cambricon states that benefiting from the continuous rise in computing power demand in the AI industry, the company has expanded its market through its competitive products and actively promoted AI application scenarios, resulting in substantial revenue growth compared to the previous year. Currently, Cambricon’s main business involves the research, design, and sales of AI core chips used in various cloud servers, edge computing devices, and terminal devices. The company’s main product lines include cloud-based products, edge products, IP licensing, and software.

According to IDC estimates, China’s intelligent computing power market is expected to grow at a compound annual growth rate of 33.9% from 2022 to 2027. By 2027, China’s intelligent computing power is projected to reach 1117.4 EFLOPS (based on FP16 calculations).

Cambricon believes that large models like ChatGPT and DeepSeek require powerful computing support from training to inference. As models become more complex, the demand for computing power also increases, potentially creating new growth opportunities in the AI chip market. With emerging technologies such as IoT, cloud computing, and big data maturing, the number of computing devices equipped with smart chips will continue to grow rapidly, whether data is stored in the cloud or at the edge.

In terms of dividends, the company plans to distribute a cash dividend of 15 RMB per 10 shares (tax included) and transfer 4.9 shares for every 10 shares from capital reserves. In 2025, it plans to pay a total cash dividend of 632 million RMB (tax included), accounting for 30.71% of net profit attributable to the parent company.

Exiting the Sci-Tech Innovation Growth Tier

Domestic Computing Power Moving Toward Value Realization

It is noteworthy that, with Cambricon’s first full-year profit, the company is also expected to be among the first batch of companies to exit the Sci-Tech Innovation Board’s Growth Tier.

On June 18, 2025, the China Securities Regulatory Commission introduced further reforms to the Sci-Tech Innovation Board with the “1+6” policy measures, including the establishment of a Growth Tier. On July 13 of the same year, the Shanghai Stock Exchange issued guidelines for the Sci-Tech Innovation Board Growth Tier, further emphasizing support and inclusiveness for “hard tech” companies. The Growth Tier is designed to support tech enterprises that have achieved significant technological breakthroughs, have broad commercial prospects, invest heavily in R&D, and are not yet profitable at the time of listing.

Industry insiders believe that the high barriers to entry, heavy R&D investment, and long development cycles characterize the chip industry. Companies typically follow a development path of “intensive R&D—revenue growth—narrowing losses—profit realization.” As a leading domestic provider of computing power, Cambricon’s profitability indicates that domestic AI computing power is entering a stage of value realization amid explosive demand.

Some analysts further suggest that in the future, domestic chip manufacturers like Cambricon may benefit from China’s token overseas expansion.

According to OpenRouter, a global AI model API aggregation platform, during the week of February 9-15, 2026, China’s model API calls reached 4.12 trillion tokens, surpassing the US models’ 2.94 trillion tokens for the first time. The following week (February 16-22), China’s model calls further increased to 5.16 trillion tokens, a 127% jump from three weeks prior, while US model calls declined to 2.7 trillion tokens, widening the gap between Chinese and US models.

In terms of market landscape, among the top five models globally from February 16-22, China occupied four spots: MiniMax M2.5, KimiK2.5 from Yuezhiyin, GLM5 from Zhipu, and DeepSeek V3.2. These four models contributed 85.7% of the total top five calls.

Guohai Securities research reports indicate that Chinese model developers are leveraging rapid technological iteration and significant cost advantages to form strong global competitiveness, driving domestic AI computing power demand into an exponential growth phase.