BTC $79,706.50 -1.93%
ETH $2,452.59 -2.31%
BNB $719.26 -0.34%
XRP $1.40 -4.23%
SOL $101.68 -3.24%
TRX $0.3318 +0.26%
DOGE $0.0847 -5.06%
ADA $0.2126 -3.94%
BCH $252.42 -2.33%
LINK $11.66 -1.16%
HYPE $85.15 +1.02%
AAVE $130.53 -2.87%
SUI $0.7551 -4.26%
XLM $0.1787 -3.73%
ZEC $1,025.08 +6.50%
BTC $79,706.50 -1.93%
ETH $2,452.59 -2.31%
BNB $719.26 -0.34%
XRP $1.40 -4.23%
SOL $101.68 -3.24%
TRX $0.3318 +0.26%
DOGE $0.0847 -5.06%
ADA $0.2126 -3.94%
BCH $252.42 -2.33%
LINK $11.66 -1.16%
HYPE $85.15 +1.02%
AAVE $130.53 -2.87%
SUI $0.7551 -4.26%
XLM $0.1787 -3.73%
ZEC $1,025.08 +6.50%

aci

All
Article
Flash

first_img Micron plans to expand HBM production capacity to approximately 100,000 wafers per month by the end of the year

According to Electronic News on September 3, Micron plans to double its high bandwidth memory (HBM) production capacity compared to last year, with a maximum addition of 60,000 wafers per month by the end of this year. Insiders say that Micron's HBM output was about 40,000 to 50,000 wafers per month last year, and it is expected to reach about 100,000 wafers per month by the end of the year, narrowing the gap with Samsung Electronics and SK Hynix.Micron is increasing procurement orders from multiple HBM equipment suppliers, with production bases mainly in Taiwan and Singapore, continuously introducing equipment. Currently, the output is mainly HBM3E 12-layer, with HBM4 12-layer accounting for about 20% to 30% at the beginning of this year, which may rise to as high as 50% by the end of the year. HBM4 12-layer will be used in NVIDIA's latest AI accelerator, Vera Rubin, and Micron has started mass production of this product since the second quarter of this year.Micron CEO Mehrotra stated in the Q3 earnings call for fiscal year 2026 in June this year that the ramp-up speed of HBM4 12-layer mass production is about twice that of HBM3E 12-layer, and cumulative revenue from HBM4 shipments has exceeded $1 billion. The industry believes that Samsung and SK Hynix each have HBM production capacities of about 150,000 to 200,000 wafers per month. Counterpoint data shows that in Q2 of this year, the global HBM market shares were SK Hynix 50%, Samsung 32%, and Micron 18%. Micron is also preparing related investments in Hiroshima, Japan.

first_img Analysis: CXMT's production capacity has peaked, and DRAM prices continue to rise sharply

Analysis indicates that China's major DRAM manufacturer CXMT's monthly wafer output has reached a peak of approximately 240,000 pieces by the end of 2025, and is expected to remain flat in 2026. Due to the tightening of U.S. export controls on advanced semiconductor equipment, especially EUV lithography machines, its capacity for expansion is limited, and substantial expansion will not occur until at least 2027, depending on the progress of the domestic equipment supply chain. According to Goldman Sachs data, CXMT's coverage of domestic DRAM demand is only about 41% in 2026 and about 50% in 2028, reflecting a structural bottleneck for many years.TrendForce data shows that traditional DRAM contract prices are expected to surge by 90%-95% quarter-on-quarter in the first quarter of 2026, followed by another increase of 58%-63% in the second quarter. Jefferies predicts that prices will continue to rise by 40%-50% and 30%-40% in the third and fourth quarters, respectively. The price increase for server DRAM is even steeper, with Samsung and SK Hynix proposing price hikes of 60%-70% to clients like Microsoft and Google in the first quarter. S&P Global expects Samsung's traditional DRAM revenue per bit to rise by 116% year-on-year to $0.79 in 2026, while Micron's ASP will increase by 54% to $1.06; Bernstein predicts that SK Hynix's DRAM gross margin could reach 92.7% in the fourth quarter of 2026.Multiple forecasts suggest that effective supply relief may not occur until the end of 2027 or even 2028.

first_img HP, ASUS, and Acer reduce production capacity in Southeast Asia, with the focus of notebook production returning to China

According to DIGITIMES, the NB supply chain has reported that brand manufacturers are significantly returning to production in China, with "returning to China for production" becoming a new trend. Brands such as HP, ASUS, and Acer have recently noticeably reduced their originally planned capacity shifts to Southeast Asia, with production focus returning to Chongqing or Kunshan in China, and increasing orders to Chinese ODM or EMS manufacturers and the proportion of private label products. This involves production lines at Quanta, Inventec's Thailand factories, and Compal, Wistron’s Vietnam factories, which may be affected by customer transfers. Lenovo, Dell, and Apple, on the other hand, remain relatively unchanged.The supply chain indicates that after the U.S. equivalent tariffs were ruled unconstitutional by the Supreme Court, the impact of substitute tariffs is smaller, and the production costs for NB in Southeast Asia are on average $9 higher per unit than in China. The gross profit margin for brand manufacturers per unit of NB is mostly between 3% to 10%. Based on a low-priced model with an average price of about $300, the $9 difference is close to or even exceeds their profit, resulting in unprofitable shipments. Local Chinese governments are also demanding increased production due to reduced tax revenues, or they will reclaim past subsidies, creating a pull for returning production.Relevant ODM personnel stated that they will respect customer decisions, and if they cannot take on a single product, they will shift production locally to servers and other product lines such as networking. Lenovo is more cautious due to its diversified layout, with limited relocation; Dell has not significantly returned due to the high proportion of U.S. government orders; Apple targets the high-end market and promotes automation, with the main reason for relocating MacBook production from Shanghai being the pandemic-related supply chain disruptions, and its plans in Vietnam remain unchanged to date.

Anthropic's flagship model is facing a downturn, with only 11% of corporate spending directed towards Fable5

On August 24, according to the Financial Times, Anthropic's U.S. corporate clients are reducing their use of its most powerful model, Fable5, opting instead for lower-priced alternative models, raising concerns about its high-investment business model. Anthropic is currently preparing for an IPO, with the market expecting its valuation to reach $2 trillion or more, potentially listing as early as September.Data from Ramp on spending by 70,000 companies shows that two months after the release of Fable5, corporate spending on this model accounted for only about 11% of Anthropic's total tool spending, and has recently stabilized. Analysts believe the main reason is that Fable5 is expensive, while older models can meet the needs of most companies; the Opus5 launched at the end of July is smaller and cheaper, with corporate spending already exceeding that of Fable5. Fable5 was withdrawn shortly after its release in early June due to intervention by the Trump administration on national security grounds, and was allowed to go back online on July 1, but demand and adoption rates remain lower than those of previous cutting-edge models.Anthropic's annualized revenue reached $65 billion last month, up from $47 billion in May, growing nearly sevenfold since the beginning of the year; the company achieved adjusted operating profit for the first time in the second quarter and has 6,000 clients with annual spending exceeding $100,000. Companies are controlling AI spending by choosing models with better cost-performance ratios, while OpenAI has seen a rebound, with annualized revenue growing 35% this quarter after the release of GPT5.6, exceeding $40 billion. The market is shifting from "chasing the strongest model" to placing greater emphasis on a comprehensive balance of model capability, price, and actual ROI.
app_icon
ChainCatcher Building the Web3 world with innovations.