2026-06-07 — views $CRDO · Credo Technology Group · Active Electrical Cables / optical DSP / silicon photonics

Credo's $437M quarter and the AEC-to-optical pivot: an interconnect leader bets the next leg on photonics

Credo Technology reported Q4 FY2026 revenue of $437.0M (+157% YoY, +7.4% QoQ) and non-GAAP EPS of $1.16 vs $1.05 expected, capping a $1.335B fiscal year. But the headline is forward-looking: management guided to 80%+ FY2027 growth and called an optical inflection, targeting

Credo’s interconnect business just tripled, and it’s pivoting again

Credo Technology, the company that built a franchise selling Active Electrical Cables (AECs) into AI racks, reported fiscal Q4 2026 results on June 1, 2026, and the numbers describe a business operating at a different scale than a year ago. Quarterly revenue hit a record $437.0 million, up 157.0% year over year and 7.4% sequentially. Non-GAAP EPS came in at $1.16, ahead of the $1.05 consensus. GAAP gross margin was 68.2% and GAAP diluted EPS was $0.88, with GAAP net income of $169.1 million.

For the full fiscal year, revenue reached $1,335.1 million, more than tripling year over year, while non-GAAP net income rose more than fivefold to $661.5 million on a 47.8% non-GAAP operating margin. The company closed the year with roughly $1.4 billion in cash and short-term investments.

Why the AEC franchise matters in AI racks

Credo’s core product is unglamorous but structurally important. As GPU clusters densify, the copper-versus-optics decision inside and between racks becomes a real cost and reliability lever. On the earnings call, management framed AECs as “the preferred solution for in-rack connectivity and for many multi-rack deployments up to 7 meters,” citing “up to 1000 times greater reliability than commodity laser-based optical modules while consuming much less power.” That is the pitch: for short reach, a smart copper cable with a DSP beats firing up an optical module that runs hotter, costs more, and fails more often.

The risk in that franchise has always been concentration. In Q4, the three largest customers were 34%, 27%, and 16% of revenue, respectively. Management says it is “deployed and high volume with five of six” major hyperscalers, which both validates the technology and underscores how few buyers move the needle. The stated path to diversification runs through additional hyperscalers and “Neo Cloud” operators, which management thinks “could be on the order of 20%” of revenue over time.

The real story: an optical inflection

The forward guidance is where this becomes an interconnect-strategy story rather than a quarterly beat. Credo guided fiscal Q1 2027 to $465M-$475M with a 67%-69% non-GAAP gross-margin range, and pointed to more than 80% year-over-year revenue growth for full fiscal 2027. The driver is a deliberate move up the stack into optics: management targets more than $600 million of optical revenue in FY2027, with optical DSPs, silicon-photonics PICs, and ZeroFlap optics each contributing over $100 million.

Metric	Figure
Q4 FY2026 revenue	$437.0M (+157% YoY, +7.4% QoQ)
Q4 non-GAAP EPS / consensus	$1.16 / $1.05
FY2026 revenue	$1,335.1M (more than tripled)
FY2026 non-GAAP net income	$661.5M (more than 5x)
Q1 FY2027 guidance	$465M-$475M
FY2027 outlook	80%+ YoY growth
FY2027 optical target	$600M+ (DSP, SiPho PIC, ZeroFlap each $100M+)

This is a company that proved out copper interconnect, banked the cash, and is now using that position to attack the optical adjacency, betting that the same hyperscaler relationships that bought AECs will buy its optical DSPs and silicon-photonics modules as reach requirements grow.

Practitioner note

For anyone tracking AI-infra interconnect, two figures are worth watching across the next two prints. First, whether the largest-customer 34% concentration eases as the fifth and sixth hyperscalers plus Neo Clouds ramp; second, whether optical revenue actually crosses the $100M-per-product thresholds management laid out, since the 80%+ FY2027 growth narrative leans heavily on that inflection landing on schedule. Mid-single-digit sequential growth implied for the first half means the optical step-up is back-half weighted, so the H1/H2 cadence is the tell.

Under-considered angle

The market reads Credo as a clean AI-interconnect beneficiary, but the AEC-to-optical pivot quietly changes what kind of company it is. Selling smart copper cables is a relatively defensible niche where Credo built a reliability and power story competitors struggled to match. Optical DSPs and silicon photonics put it into a far more crowded room, where incumbents in the optical-module and DSP supply chain already compete hard on price and where the customer set partly overlaps with vertically integrating hyperscalers building their own silicon. The optimistic case is that AEC relationships pull optics through; the under-priced risk is that optics is simply a lower-moat, more-contested business than the copper franchise that earned Credo its multiple, and that mixing the two pressures the 68% gross margins that make today’s story look so clean. The guidance band of 67%-69% is the first hint that management knows it.