ASML Reports €8.8B in Q1 Revenue and Raises Full-Year Guidance — AI Memory Demand Now Drives 51% of Tool Sales
ASML beat Q1 2026 estimates with €8.8 billion in net sales and €2.8 billion net income, then raised its full-year revenue guidance to €36–40 billion. AI-driven memory chip demand hit a record 51% of quarterly tool bookings.
ASML reported first-quarter 2026 results on April 15th that beat analyst estimates across the board and prompted a meaningful guidance raise — confirming that AI infrastructure demand continues to pull semiconductor equipment at a pace that is outrunning the industry’s ability to supply it.
The numbers
Q1 net sales hit €8.8 billion against an €8.5 billion consensus estimate. Net income came in at €2.8 billion versus a €2.54 billion estimate. Gross margin held at 53.0%. Clean beats everywhere.
ASML raised its full-year 2026 revenue guidance from €34–39 billion to €36–40 billion, with a new midpoint of €38 billion. For Q2, ASML guided €8.4–9.0 billion in net sales — the €8.7 billion midpoint lands slightly below the €9.04 billion LSEG consensus, a modest softening that pushed shares lower in early Amsterdam trading despite the overall strength.
Memory is the story
The sharpest data point in the report: memory chips accounted for 51% of new tool bookings in Q1 2026, up from 30% in Q4 2025. That’s a 21-percentage-point swing in a single quarter.
The driver is straightforward: South Korean memory manufacturers Samsung and SK Hynix are aggressively ramping capacity to meet AI-driven HBM demand. HBM — High Bandwidth Memory — is the stacked memory architecture that sits directly on high-end AI accelerators. Every Nvidia H200 uses it. Every Google TPU pod depends on it. As AI training clusters scale, HBM demand scales at the same rate.
South Korean customers represented 45% of ASML’s Q1 revenue. Taiwan — home to TSMC — came in at 23%. When Korea is buying more EUV machines than Taiwan, you’re watching a memory supercycle in real time.
EUV production targets
ASML plans to ship at least 60 EUV Low NA systems in 2026, rising to at least 80 in 2027 if customer demand supports the ramp. EUV lithography is the only manufacturing technology capable of printing sub-5nm features — it’s a hard dependency for leading-edge AI accelerators, advanced smartphone SoCs, and HBM3E memory.
High NA EUV, the next-generation system that enables even finer resolution features, remains on track for volume production ramp but is not yet a significant 2026 revenue contributor.
The read
ASML is effectively a monopoly on leading-edge chip manufacturing equipment. When ASML’s tool mix shifts heavily toward memory, it maps where the supply build-out is actually happening: not just in AI logic chips like GPUs, but across the complete compute stack — including the memory that makes large-scale AI inference efficient and economically viable. The cycle is not peaking. The guidance raise says as much.