TSMC Q4 Earnings 2025: Profit Surges 35% on Relentless AI Demand
TSMC reported a 35% surge in Q4 2025 profit, hitting a fresh record of NT$505.74 billion driven by strong AI chip demand from Nvidia and AMD.
The AI boom isn't just a trend—it's a gold mine for the world's largest contract chipmaker. TSMC has just shattered expectations, reporting year-over-year profit growth for the eighth consecutive quarter as the hunger for advanced AI chips shows no signs of slowing down.
TSMC Q4 Earnings 2025: Financial Records Broken
According to official reports, TSMC posted a 35% jump in fourth-quarter net income, reaching a record NT$505.74 billion (approx. $15.6 billion). This comfortably beat LSEG SmartEstimates of NT$478.37 billion.
| Metric | Actual Results | Market Expectation |
|---|---|---|
| Revenue | NT$1.046 trillion | NT$1.034 trillion |
| Net Income | NT$505.74 billion | NT$478.37 billion |
| Revenue Growth (YoY) | 20.5% | - |
The 7nm and Below Dominance
The company's dominance in high-performance computing (HPC) is undeniable. Advanced chips measuring 7-nanometer or smaller accounted for a massive 77% of total wafer revenue. This segment includes critical processors for giants like Nvidia and AMD, who rely on TSMC's cutting-edge efficiency to power AI servers.
Looking Ahead to 2026
Analysts predict that 2026 will be another breakout year. With the ongoing expansion of 2nm capacity and the growth of advanced packaging solutions, TSMC is well-positioned to maintain its lead. However, some risks remain, as a persistent memory shortage could dampen the recovery of the consumer electronics sector, including smartphones and PCs.
Authors
Related Articles
SK hynix posted its first 100-trillion-won quarter, with an 80% operating margin. Micron's GAAP margin was 1.0pp higher; TSMC and Samsung were in the 50% range.
Five stories from late September, traced to where the paper trail stops: Korea's export record, Nvidia's Hugging Face 8-K, a US-China AI channel, the UN on AI, and rising births.
Huawei's CANN is open source under a license barring software for non-Ascend chips. Inside China's second CUDA, and the two links still outside it: pre-training compute and 5nm capacity.
Nvidia shipped roughly a billion RISC-V cores in 2024, then announced it would run CUDA on the open standard. We break down how royalty-free instruction sets and open software stacks are trying to route around CUDA's lock-in. Part 2 of the Semiconductor Sovereignty series.
Thoughts
Share your thoughts on this article
Sign in to join the conversation