I now have enough for a comprehensive preview. Let me pull one more data point — quick check on Advantest's recent commentary/TAM (competitor read-through) since it was referenced in the Q1 call.# Teradyne (TER) Q2 2026 Earnings Preview Report Date: Tuesday, July 28, 2026 (after market close) | Call: Wednesday, July 29, 8:30 a.m. ET
Teradyne enters this print riding one of the strongest AI-demand narratives in the semiconductor capital equipment space, but it does so against a backdrop of extreme stock volatility and a broader chip-sector risk-off move. TER shares closed at a record $483.84 on June 30, then whipsawed lower through July — closing at $334.81 on July 27, roughly 31% below that peak, as the semiconductor complex was rattled late last week by reports of Chinese domestic lithography production and a blockbuster Shanghai IPO from memory maker CXMT, which hit ASML, SK Hynix, and other chip-supply-chain names. Sentiment among sell-side desks heading into the print has been described as bullish on fundamentals but nervous on positioning: expectations call for sales up 87% to roughly $1.21 billion (semi test +112% to $1.041 billion), gross margin near 58.5%, and EPS up over 260% to about $2.05, with the market signaling that "any disappointment will be punished much more than an upside print will be rewarded."
Notably, Teradyne's own history this cycle argues for caution about "sell the news" risk: despite reporting a blowout, record-beating Q1 (revenue and EPS both above the high end of guidance), TER stock fell roughly 19% the very next day, as investors focused on lumpy back-half guidance and a rich valuation rather than the beat itself.
Teradyne's first quarter was the strongest in company history. Management noted that with revenue of approximately $1.3 billion and non-GAAP EPS of $2.56, Teradyne delivered record results in the first quarter of 2026, and that this new record comes from durable AI demand drivers and the continuing acceleration of our Wafer to AI data center strategy. Critically, AI-related demand accounted for nearly 70% of our revenue, up from about 60% in Q4 of 2025 — a rapid concentration shift investors should track closely into Q2.
By segment: SemiTest revenue broke through the $1 billion mark for the first time (SoC $883M, Memory $203M, IST $27M), Robotics posted its fourth consecutive quarter of sequential growth, and Product Test grew on defense/aerospace and board-test strength. Gross margin hit a record 60.9%, aided by mix and one-time operational benefits, and operating margin reached 37.5%.
Teradyne's own guidance, issued alongside Q1 results, calls for: - Revenue: $1.15B–$1.25B (implying ~84% YoY growth at the midpoint, a sequential decline from Q1's $1.28B) - Non-GAAP EPS: $1.86–$2.15 - Gross margin: 58%–59% (down from Q1's 60.9%, as one-time benefits normalize) - OpEx: ~27%–28% of revenue - Non-GAAP operating margin: 30%–32%
Management was explicit that roughly half of the sequential gross-margin step-down is due to the roll-off of one-time operational benefits, with the rest reflecting normal mix normalization, and cautioned that margins move around a bit... they are lumpy like our revenue... up to 400 basis points within a year even though the year-on-year range is tighter.
On timing, CFO Michelle Turner widened (rather than narrowed) the first-half revenue mix guide to 55%–60% of full-year revenue, versus a firmer 60% point estimate given back in January — a subtle but important signal that order lumpiness/timing risk into the back half has increased, even as demand signals themselves have strengthened.
1. AI "Wave 2" and Merchant GPU Ramp. Teradyne frames the AI opportunity as three overlapping waves — general-purpose data center build-out (2025), inference-optimized compute silicon (2026), and eventual edge/physical AI (later). The company disclosed it received its first multisystem production test orders for merchant GPU in Q1, with systems expected to ship and enter production in Q2, and management indicated ~$50 million of merchant GPU revenue line-of-sight for the full year, though second-half visibility there is "quite limited." Investors should listen for early commentary on conversion pace, since management said achieving even a mid-teens/low-double-digit share of merchant GPU testing would be enough to hit the long-term target model.
2. Memory (HBM/DRAM) Strength — A Potential Upside Swing Factor. Management flagged that memory test demand is running stronger than its January view, aided by both HBM and DRAM AI-driven demand, and expects to gain low-single-digit share in a memory TAM headed for solid growth. Notably, Greg Smith called memory the segment most likely to skew more second-half weighted — meaning strength here could push full-year mix toward the low end (55%) of the first-half revenue range, which would actually be a bullish signal. Recent read-throughs (Micron's blowout quarter that sparked TER's June 25 record-high rally, plus Broadcom's newly expanded $200B+ memory/ASIC collaboration with Samsung and Nvidia's reported ~$500B SK Hynix supply agreement) all point to a tightening global memory market that could benefit Teradyne's memory-test franchise.
3. Silicon Photonics / Co-Packaged Optics (CPO) — Early But Real Optionality. Teradyne launched Photon100, its new CPO/silicon-photonics test platform, sizing the mid-term TAM opportunity at $300–$700 million per year, while cautioning this is "very early days" (2026 company-level CPO revenue guided around roughly $100 million) and that efficiency gains could rise ~10x over the next couple of years even as the market scales.
4. New Products/M&A extending the "wafer to data center" strategy. Teradyne closed its MultiLane Test Products JV (April 8) and the TestInsight acquisition (mid-April), together using ~$165 million of cash, funded via revolver — both aimed at high-speed I/O/data-center interconnect test and design-to-test software capabilities. The new Omnyx board-test platform targets defect detection in AI server board/tray assemblies.
5. Auto/Industrial — Data Center Crowding Out Legacy Recovery. Auto/industrial revenue nearly doubled sequentially in Q1, but almost entirely from power-management demand tied to AI data center buildouts — with 46% of segment revenue now coming from data-center devices rather than traditional auto/industrial. Management said broader auto/industrial capital equipment demand has not yet meaningfully recovered despite reduced customer inventories.
6. Robotics — Consistent but Still "Prove It." Robotics grew 32% YoY and delivered a fourth straight quarter of sequential growth (unusual since Q1 is seasonally weak), with AI-related robotics revenue reaching 15% of segment sales. Management remains deliberately conservative on extrapolating robotics momentum until "we get a little bit further down the road."
7. Competitive/TAM Backdrop. Rival Advantest — which controls the lion's share of the SoC tester market — has been guiding to a CY2026 SoC tester TAM of roughly $8.5–9.5 billion (~30% YoY growth) and a memory tester TAM of $2.2–2.7 billion (~20% growth), alongside its own FY2026 revenue guide of ~$9 billion. Teradyne notably declined to publish its own 2026 TAM estimate on the Q1 call, with CEO Greg Smith saying management doesn't "feel confident enough in our forecast to share" a number, given how far off both companies' 2025 TAM estimates were a year ago. Watch for whether Teradyne offers more TAM specificity this quarter given rising investor scrutiny on this point.
Beyond the Q2 number, the more important large-picture question is whether management reiterates, raises, or hedges its full-year target model of ~$6 billion in revenue and $9.50–$11.00 in non-GAAP EPS. As of Q1, leadership reaffirmed confidence in that trajectory and dubbed 2026 "the year of execution." Any commentary suggesting slippage — particularly around VIP/custom-compute program timing sliding into 2027, or continued caution on H2 visibility — could be the key swing factor for the stock reaction, more so than the Q2 print itself.
Teradyne heads into Q2 earnings with genuinely strong underlying AI-driven demand — record Q1 results, a widening AI product portfolio (SemiTest, Robotics, Product Test all growing), and multiple emerging growth vectors (merchant GPU, CPO, HBM/memory, HDD/IST). The guidance itself ($1.15–$1.25B revenue, $1.86–$2.15 EPS) is not particularly demanding relative to the underlying growth story, but the stock's history this cycle (a 19% drop after beating and raising in Q1) shows that investors are laser-focused on second-half visibility, TAM commentary, and reaffirmation of the $6B/$9.50–$11 EPS target model rather than the headline beat/miss. Given the recent broad chip-sector volatility, expect an outsized stock reaction in either direction.