Camtek’s beat was small, but the mix signal was not: advanced packaging is still absorbing inspection capacity

The actionable read from this quarter is not that Camtek surprised the model materially, because it did not. What was priced in was a record-ish quarter near consensus, with revenue expected at $125.1 million and EPS expected at $0.80; what actually surprised was narrow, at actual revenue of $126.0 million and actual EPS of $0.82, or +0.7% and +2.1% versus Street. The market’s likely mistake is to file that as a fully de-risked, fully valued event while ignoring the composition of the $126.0 million. Rafi Amit framed the reported basis directly: “Q3 revenues reached a record $126 million, reflecting over 12% growth year-over-year.” That wording matters because management is not leaning on bookings language or a recovery narrative; it is pointing to realized revenue at a record level, while the business is already carrying a mix that is concentrated in advanced packaging rather than broad wafer-fab capex.

That distinction matters because Camtek’s top-line history now shows deceleration without demand deterioration. Revenue moved from $112.3 million in Q3 FY2024 to $126.0 million in Q3 FY2025, with reported revenue YoY slowing from +39.6% in Q3 FY2024 to +12.2% in Q3 FY2025. Sequentially, the company went from +9.5% in Q3 FY2024 to +2.2% in Q3 FY2025. Those numbers are slower, but they sit on a materially higher base: revenue was $80.5 million in Q3 FY2023, $112.3 million in Q3 FY2024, and $126.0 million in Q3 FY2025. A market that expected a straight continuation of the +39.6% year-over-year phase will be disappointed; a market underwriting advanced packaging inspection penetration should focus on the fact that revenue still expanded at +12.2% while gross margin stayed at 50.0% in the quarterly history and management discussed 51.5% gross margin on its own reported call basis. The print says the cycle has shifted from explosive adoption to monetization, not from growth to exhaustion.

The margin path reinforces that interpretation because the company did not buy the revenue beat with mix dilution. Gross margin in the quarterly history was 49.7% in Q3 FY2024, 49.6% in Q4 FY2024, 51.0% in Q1 FY2025, 50.8% in Q2 FY2025, and 50.0% in Q3 FY2025. That is not a collapse as revenue growth slowed from +20.2% YoY in Q2 FY2025 to +12.2% YoY in Q3 FY2025; it is a plateau around the 50.0% level. On the call basis, management gave gross profit of $65 million and operating profit of $37.6 million, with operating expenses of $27.2 million compared with $26.6 million in the previous quarter. The second-order point is that Camtek is absorbing higher operating expenses without losing operating profit momentum, since operating profit was $37.6 million in Q3 FY2025 compared with $37.4 million in the second quarter. If the next leg of growth comes from new models rather than lower-value inspection, the margin signal this quarter supports incremental revenue quality.

The EPS optics need care because the Street-comparison basis and the company’s GAAP history do not tell the same story. The print basis was $0.82 versus $0.80, a +2.1% surprise, and Moshe Eisenberg stated that “Net income for the third quarter of 2025 was $40.9 million or $0.82 per diluted share.” In the quarterly history, diluted EPS for Q3 FY2025 is -$1.16, which reflects the one-time GAAP effect management described separately: a tax asset was created in the amount of $12.3 million, which led to a onetime GAAP loss of $89 million net. Those are different reporting bases, and the investment conclusion should not mix them. The operating read is that the company produced $37.6 million of operating profit and $6.5 million of financial income in the quarter, while the GAAP EPS line was distorted by the convertible-note transaction. For PMs, the dilution and capital-structure mechanics are the real issue to model, not the GAAP loss headline.

That capital-structure issue is where the market has a legitimate reason to withhold full credit, but the data do not support treating it as demand weakness. Cash and cash equivalents, including short- and long-term deposits and marketable securities, were $794 million as of September 30, 2025, and management also referenced approximately $800 million including the additional cash generated by a $500 million convertible notes offering in Q3. The company repurchased for $267 million existing convertible notes with a balance sheet value of $167 million, producing positive net cash flow from this activity of $219 million. The share count effect is explicit: total diluted number of shares as of the end of the third quarter was $50.3 million, and in the next quarter it is expected to be around 51 million shares as the convertible notes apply to the full quarter. That means Q4 EPS estimates must absorb a higher denominator, but the balance sheet also gives Camtek capacity to pursue organic growth and potential inorganic activity without needing an equity raise at the current point in the cycle.

The guide is the one place where the data conflict enough to demand a hedge, and the hedge is specific. Rafi Amit guided Q4 revenue to “around $125 million,” representing annual revenue of $495 million and growth of 15% over 2024, while Moshe Eisenberg later said, “we expect revenues of around $127 million in the fourth quarter.” The quarterly history shows Q4 FY2025 revenue at $128.1 million, with +1.7% QoQ and +9.2% YoY. Those are not identical figures, so the defensible interpretation is a Q4 revenue band centered in the mid-to-high $120 million range rather than a single-point acceleration thesis. The market may have wanted a cleaner raise after a $126.0 million print, but a Q4 level around $125 million or around $127 million still implies that the business is holding record quarterly revenue rather than mean-reverting toward the $117.3 million level seen in Q4 FY2024. The key disappointment is not demand falling short; it is that management did not convert the Q3 beat into an unmistakably higher Q4 commitment.

The mix commentary explains why that guide can still be enough. High-performance computing applications contributed approximately 45% of total revenue, while other advanced packaging applications accounted for about 25%. Ramy Langer put the broader split at around 70% advanced packaging, with 50% HPC and another 20% conventional advanced packaging. The exact split differs by framing, but both versions point to the same investment fact: the business is no longer primarily levered to undifferentiated inspection demand. Management also said new models are expected to contribute approximately 30% of revenue in 2025, with an even larger share expected next year. That is the most important non-consensus detail in the call, because it links current revenue durability to a product transition rather than just to customer capex timing. If approximately 30% of 2025 revenue is coming from new models, the company is refreshing its installed value proposition while revenue is already at $126.0 million.

That mix has direct read-throughs for the semiconductor supply chain. For TSMC, listed as a Camtek equipment customer, a quarter in which HPC applications contributed approximately 45% of Camtek revenue suggests inspection intensity remains tied to leading-edge compute packaging rather than only wafer starts. For ASE Group, listed for advanced packaging inspection, the relevant number is the approximately 25% contribution from other advanced packaging applications, because it indicates demand beyond the HPC bucket is still material. Samsung and SK Hynix are not linkable here, but the read-through is equally concrete: Samsung’s wafer and packaging inspection exposure and SK Hynix’s HBM packaging inspection exposure sit inside a Camtek mix where advanced packaging is around 70% of revenue and Asia was 93% of quarterly revenue. The geographic number matters because any slowdown or pushout among Asian OSAT, foundry, memory, or packaging customers would show up quickly in Camtek’s revenue trajectory; this quarter, revenue still reached $126 million on management’s call basis.

The peer comparison also argues that Camtek’s issue is scale, not category relevance. In the process-control peer set, KLAC reported $3,415.1 million of revenue, 61.1% gross margin, and +11.5% revenue YoY, while Camtek printed $126.0 million and +12.2% revenue YoY with 50.0% gross margin in the quarterly history. ONTO reported $291.9 million of revenue, 50.1% gross margin, and +9.5% revenue YoY; NVMI reported $235.3 million of revenue, 57.7% gross margin, and +10.3% revenue YoY. Camtek’s growth rate is not out of line with the process-control group, and its gross margin is essentially adjacent to ONTO’s 50.1%, though below NVMI’s 57.7% and KLAC’s 61.1%. The variant point is that Camtek is being judged on a decelerating small-cap growth curve while its margin and growth profile are already comparable to named process-control peers at larger revenue bases.

The call delivery was better than the guidance headline suggests, but not clean enough to ignore the Q4 ambiguity. The tone history shows Q3 FY2025 sentiment at 0.38 versus 0.28 in Q2 FY2025, prepared_sentiment at 0.66 versus 0.61, and ai_optimism at 0.39 versus 0.27. Those improved, but guidance_tone fell to 0.03 from 0.35, while uncertainty rose to 66.5 from 54.1 and qa_evasiveness moved to 11.7 from -21.8. That combination fits the transcript: management was more positive in describing the quarter and mix, but less forceful in translating that into Q4. The tone series is therefore a check on the thesis rather than a victory lap. It supports the idea that the business mix is improving, while warning that near-term orders are not being messaged with the same clarity as the product and end-market narrative.

That tension is exactly why the stock debate should center on confirmation points rather than on the modest beat. The bull case is confirmed if Q4 revenue lands at least around $125 million, or closer to the around $127 million comment, while gross margin stays near the 50.0% level shown for Q3 FY2025 and Q4 FY2025 in the quarterly history. It is also confirmed if advanced packaging remains around 70% of revenue, HPC remains near approximately 45% or the 50% framing used by Ramy Langer, and new models track toward approximately 30% of 2025 revenue with an even larger share next year. It breaks if Q4 revenue falls below the around $125 million guide, if gross margin moves materially below the 50.0% level, or if Asia’s 93% concentration begins to translate into receivables or inventory stress rather than revenue. On working capital, the next check is whether inventory remains improved from $149 million to $142 million and whether accounts receivable stay near $112 million and 81 days outstanding. The print did not give investors a large numerical beat; it gave them a narrow beat attached to a better mix than the headline deceleration implies. That is enough to stay constructive, but only if the next quarter proves that the mid-$120 million revenue level is a floor, not the top of the advanced packaging inspection cycle.