Summary of Key Points
In the past two weeks, several significant events have taken place in the memory chip industry: Micron Technology released its strongest financial report ever (with revenue soaring by 346% and a gross margin exceeding 84%). However, American analysis firms predict that China's ChangXin Memory Technologies will overtake Micron to become the world's third-largest DRAM supplier by the end of 2026. Meanwhile, Google is evaluating the possibility of purchasing DRAM from ChangXin, which has also secured 29.5 billion yuan in funds from its IPO on the STAR Market. Behind these developments lies a competitive landscape involving China, South Korea, and the United States in the memory chip sector. ChangXin is rapidly rising in the general-purpose DRAM market, but it still lags behind in HBM (the core storage technology for AI). Micron's market share in China is declining, providing an opportunity for ChangXin to fill the gap. The varying progress of Chinese memory companies across different segments indicates that its position as "world third" reflects both cyclical advantages and genuine improvements.
I. Three Different Markets: China's Roles in Each Segment
The memory chip industry consists of three distinct markets, each with a unique competitive landscape for Chinese companies:
1. HBM Market (the most high-end segment for AI): HBM is essential for GPUs, which require fast processing but cannot function without data from HBM. This market is dominated by South Korea's SK Hynix (58%) and the United States' Samsung (21%), with Micron also selling out its production capacity (21%). China currently has no significant presence in this high-end segment.
2. General-Purpose DRAM Market (the largest market): This includes DDR/LPDDR memory used in smartphones, computers, and servers, traditionally dominated by Samsung (38%), SK Hynix (29%), and Micron (22%). ChangXin ranks fourth with 8% of the market share but has seen impressive growth—revenue increased by 719% and profits by 1688% in Q1 2026. The reason for this is that the three giants have shifted their production capacity to HBM (one HBM chip can replace three conventional DDR chips), leading to a shortage of general-purpose DRAM. Companies like Alibaba Cloud, Tencent, and Xiaomi are using ChangXin's products, and Google is also considering purchasing from them.
3. NAND Market (flash memory for storage): This includes storage in smartphones and SSDs, led by Samsung (29%) and SK Hynix (18%). China's Yangtze Memory Technologies has entered the top three with a 13% market share, thanks to its own research and development efforts, and could potentially overtake Micron after further expansion. However, NAND profits are lower, and its connection to AI applications is weaker, resulting in smaller price increases compared to DRAM.
In summary, China is still in the early stages of HBM development, competing fiercely for a share in the general-purpose DRAM market, and has already made significant progress in the NAND segment.
II. ChangXin's Profitability: Cyclical Benefits or Real Competence?
ChangXin's financial results seem impressive, but a closer look reveals some nuances:
- Cyclical Factors: The 80% increase in profits came mainly from price hikes (DRAM contract prices rose by 93%-98% in Q1 2026), not from increased sales volume (which only increased by 11%). Additionally, ChangXin's DRAM production costs are 30% higher than those of the three giants, meaning its profits rely on price increases. If prices fall, its cost disadvantage will become evident. Most of its products are standard DRAM, and it does not yet have access to high-profit server DRAM or HBM markets.
- Real Competence: ChangXin's yield rates are improving (nearly 90% for DDR5), and its production capacity is the largest in the world (with an additional 85 kilowatt-hours per month planned for 2026). Its customer base is expanding, including leading domestic companies and Google. Importantly, ChangXin did not start from scratch; it acquired 7,000 patents from German company Qimonda in 2015, which was once the world's second-largest DRAM manufacturer. This, combined with ten years of sustained investment in Hefei, has helped it overcome previous losses (a total of 36.6 billion yuan in losses).
In conclusion, while ChangXin's current success is partly due to cyclical factors, it also reflects genuine technological and operational improvements.
III. Micron vs. ChangXin: The Competition in the Chinese Market
The relationship between Micron and ChangXin is one of mutual competition:
- Micron's Decline: In 2023, China's Cyberspace Administration reviewed Micron products and required that critical infrastructure stop using them, causing Micron's revenue share in China to drop from a high point to 7%. Micron has also laid off employees in its Shanghai and Shenzhen teams, and its market share in the mobile segment has shrunk significantly.
- ChangXin's Rise: The market space vacated by Micron has largely been taken over by ChangXin, with its share of China's mobile DRAM market growing rapidly. Companies that previously used Micron products, such as Alibaba Cloud and ByteDance, have switched to ChangXin. Even Google (a foreign company) is turning to ChangXin due to the shortage of general-purpose DRAM caused by the shift in production capacity towards HBM.
It's important to note that the three giants are only temporarily giving up control of the general-purpose DRAM market; they can return at any time if HBM supply becomes more stable.
IV. The Meaning of “World Third”: Short-Term Advantages and Long-Term Potential
ChangXin's position as "world third" should not be overestimated:
- Short Term: Its profits are largely due to price increases, and its HBM yield rates are still too low (25%) for commercial use. Lacking access to EUV lithography machines (essential for advanced chip production), it is still some way from achieving true technological leadership.
- Long Term: ChangXin's progress over the past decade, from a non-player to the fourth-largest supplier in the world, indicates significant growth in capacity, customer base, and technology. The 29.5 billion yuan raised through its IPO will be used for expansion and technology upgrades, with Google's order serving as a testament to its quality. More importantly, China has moved from being a bystander in the memory chip industry to becoming a key player in a tripartite competition.
If ChangXin can break through in HBM (with a projected 12% market share by 2028), it will truly compete on an equal footing with South Korea and the United States.
Final Summary
ChangXin's achievement of third place globally marks a decade of significant progress for China's memory industry. However, it has not yet surpassed Micron in terms of overall strength. Its current success is partly due to cyclical factors, but long-term success will depend on its ability to overcome challenges in HBM and EUV technology. For now, Chinese memory companies are no longer mere observers; they are actively shaping the industry landscape.