Is CAMT a Buy or a Sell? The Bull and Bear Case (2026)

That depends on which case you find more convincing, and both are on this page. The bull case rests on HBM and the memory stack, with revenue (ttm) at ~$499M (FY2025 ~$496M, up ~16%). The bear case rests on customer concentration is real: a handful of memory makers, foundries and OSATs decide most of the year, and a single pushed order can move a quarter. Analysts covering it are spread from $152.00 to $200.00, which is itself a signal that this is genuinely contested. If you believe the thesis, the real questions become sizing and overlap rather than timing. Walnut is not an investment adviser and this is not a recommendation.

Nobody can answer that for you, and the honest version of the question is narrower: has anything changed in the reason you bought it? The bear case on this page is the place to check. Customer concentration is real: a handful of memory makers, foundries and OSATs decide most of the year, and a single pushed order can move a quarter. If that risk is what you were worried about and it is now playing out, that is a real signal. If the price simply fell while the thesis held, that is a different situation entirely. The most pessimistic published target is $152.00, +38.0% from the $110.15 price, which is one analyst's downside case rather than a floor. Walnut is not an investment adviser.

HBM and the memory stack. Every additional layer in a high-bandwidth memory stack adds inspection and measurement steps, and custom HBM designs that put a logic die beneath the DRAM stack add more still. The most optimistic analyst target on CAMT is $200.00, +81.6% from the $110.15 price. That figure is only reachable if this thesis works close to its best case.

Customer concentration is real: a handful of memory makers, foundries and OSATs decide most of the year, and a single pushed order can move a quarter. Geographic concentration compounds it, with the large majority of sales into Asia Pacific and roughly a third historically into China, exposing the company to export controls and tariff policy on both sides. Growth is lumpy rather than smooth, as the flat first quarter of 2026 followed by a jump to about $180 million of guided Q4 revenue shows, and the 2027 outlook of more than $800 million assumes customers keep taking tools on schedule. Valuation leaves little margin for error even after the pullback, with the shares near 12 times trailing sales and roughly 26 times forward earnings as of October 2026. The share price has also proven volatile: it fell about 16 percent in the month to October 16, 2026 and sits roughly 42 percent below its 52-week high of about $216, despite raised guidance. Camtek is also a much smaller company than KLA and competes against a rival with far greater scale in tools, software and service. The most pessimistic published target is $152.00, +38.0% from the current price, which is roughly what the stock is worth if these risks bite rather than the drivers.

Camtek makes inspection and metrology systems used in chip packaging, giving it direct exposure to the advanced-packaging spending that AI accelerators require.

Decide that in advance, because it is far harder to think clearly once a position is moving against you. The concrete tripwires here: the driver behind the bull case (HBM and the memory stack) stalling in the reported numbers rather than in the narrative, the risk above (customer concentration is real: a handful of memory makers, foundries and OSATs decide most of the year, and a single pushed order can move a quarter) turning from a possibility into a reported fact, or the position growing large enough that a bad outcome would matter to your whole portfolio regardless of who is right.

Optical inspection and metrology machines used during chip manufacturing and packaging. The Eagle systems handle two-dimensional defect inspection, scanning wafers and packages for flaws, while the Hawk line performs three-dimensional metrology, measuring the height and shape of features such as copper pillars, bumps and through-silicon vias. Customers buy them because catching a defect before an expensive AI package is assembled saves far more than the tool costs.

AI accelerators are built by stacking dies together, using techniques like CoWoS packaging and high-bandwidth memory. Each added layer and interconnect creates more places for defects and more dimensions that must be measured, which raises the number of inspection steps per part. Camtek says roughly 70 percent of its revenue is now AI-related, so the stock tends to move with AI capacity announcements from memory makers, foundries and packaging houses.

Equipment revenue depends on when tools ship, not when demand appears. First quarter 2026 revenue of about $122 million was up only around 2.5 percent from the year-ago quarter after a 16 percent growth year in 2025, because customers were in a digestion period. Management simultaneously reported record order intake and guided the second half of 2026 to more than 25 percent above the first half, which is the shape of a timing gap rather than a demand gap. That is how it played out: Q2 revenue rose to about $133 million and preliminary Q3 revenue came in at a record $160 million to $161 million, with Q4 guided to about $180 million.