Cohu secures multi-unit order for Eclipse test platform from U.S. chip maker

Cohu reported Q4 2025 EPS of -$0.15 versus a $0.06 consensus (a -350% surprise) while revenue slightly beat at $122.2M vs. $122.1M. The company received a multi-unit order for its Eclipse AI/HPC test platform (its second customer) and sees HPC-segment revenue toward the upper end of a $65M–$80M internal projection; test-cell utilization improved to 78%. Analysts reacted variably: Needham raised its PT to $33 (Buy) and Evercore initiated Outperform with a $35 PT, while InvestingPro flags the stock as currently overvalued despite analysts projecting profitability this year.

Analysis

Cohu’s product set positions it to capture incremental ASPs from the highest-power AI/HPC devices where socket force and thermal control are differentiators, creating a wedge versus generic handler suppliers. Second-order beneficiaries include niche thermal-subsystem vendors and OSATs that outsource high-power validation; these suppliers will see shorter lead times but higher order stickiness and aftermarket service revenue over multiple years.

Key catalysts are installation cadence and utilization inflection: visible multi-unit deployments converting into steady service/consumables revenue within 3–12 months will materially de-risk current valuation assumptions, while a slip in AI accelerator design wins or a foundry/IDM decision to internalize test capability would reverse the re-rating quickly. Near-term volatility (days–weeks) will be driven by analyst notes and bookings disclosure; structural revenue realization plays out over 6–24 months as customer ramps and qualification cycles conclude.

Consensus excitement around AI exposure likely underweights margin and working-capital dilution risks tied to rapid scale-up (tool install costs, one-time inventory adjustments), so upside is not pure free cash flow yet. Tactical exposure via time-limited options or paired trades that hedge sector beta captures most upside while protecting against secular softness if AI capex growth moderates over the next 12 months.

Key Decisions for Investors

• Long COHU (equity) on a disciplined entry: accumulate on 8–12% pullbacks versus current levels with a 12-month time horizon; target 30–40% upside if utilization milestones and multi-unit deployments become recurring, stop-loss at 12% to limit idiosyncratic execution risk.
• Buy a 9–12 month COHU call spread to limit downside and retain upside: structure as buy-to-open 1x long Jan- or Oct- LEAP call and sell a nearer strike to finance (~cost basis ~30–40% of outright LEAP). Expect 2–3x upside if adoption scales; max loss = premium paid.
• Pair trade to isolate product-share gains: long COHU / short TER (1:1 notional) to express share shift toward high-force/thermal test architectures while hedging broader semicap cyclicality; unwind if sector breadth turns positive and both report above-trend bookings.
• Event hedge: purchase short-dated protective puts (30–60 days) around next earnings/booking cadence if you hold stock outright — cost-effective insurance given high headline risk from analyst notes and order disclosures.