Why Did Cerebras Stock Price Drop 16% After Earnings? AI Growth Was Not Enough

 

A company reported gross margin above its own prior guidance range, raised full-year revenue and margin guidance, and its stock fell roughly 16%. Understanding that outcome requires starting with its valuation multiple.

 

Cerebras Systems reported second quarter 2026 results after the US close on August 12. Per Investing.com's earnings call summary, the stock finished the regular session at $261.76, up 11.5%, before falling 15.66% after hours to $221.01. Different sources put the decline between 14% and 17%, with the variation reflecting measurement timing.

 

The immediate trigger was one number: revenue of $180.11 million against an LSEG-compiled analyst estimate of $194.23 million. But that figure is not the same measure as the $209.9 million of core revenue the company emphasised, and core revenue actually came in above the roughly $194 million core revenue guidance the company had issued. What turned this report into a failure in the market's eyes is a background statistic noted in GuruFocus's summary: the stock's price-to-sales ratio stands at an elevated 291.5 times, well above its historical median of approximately 238.6 times and the semiconductor industry average.

 

 

 

The stock closed the regular session at $261.76, up 11.5%, then fell 15.66% after hours to $221.01, with sources reporting declines between 14% and 17%.

 

Reported revenue was $180.11 million, up 74.3% year over year, below the $194.23 million LSEG consensus, while the core revenue the company highlighted on its call was $209.9 million, up 103% and a record.

 

Core gross margin was 40.6%, up from 29.6% a year earlier and above the company's own 36% to 38% second quarter guidance, but down from 46.5% in the first quarter.

 

Core operating margin was negative 16%, improved from negative 42% a year earlier, while adjusted loss per share of $2.98 compared with a consensus estimate of $0.18.

 

Full-year core revenue guidance rose to $880 million to $890 million from $855 million to $865 million, and full-year core gross margin guidance rose to 41% to 43% from 38% to 41%.

 

Third quarter guidance calls for core revenue of $214 million to $216 million, core gross margin of 38% to 40% and core operating margin of negative 25% to negative 23%.

 

Remaining performance obligations reached $25.4 billion as of June 30, with $8.6 billion in cash, cash equivalents, restricted cash and marketable securities plus an unused $850 million revolving credit facility.

 

The stock trades near 291.5 times sales against a historical median of roughly 238.6 times, after a May IPO priced at $185 that raised $6.4 billion, the largest semiconductor offering on record.

 

 

 

This is the most misread element of the report and the first thing that needs clarifying. Investing.com's summary states explicitly that revenue was reported at $180.11 million in the earnings data while the company's call highlighted core revenue of $209.9 million.

 

The two figures reflect different measures. Reuters coverage published via Investing.com compared $180.11 million with the $194.23 million analysts expected per LSEG and concluded results fell short. Meanwhile the company's prior quarter filing with the SEC shows second quarter core revenue guidance of approximately $194.0 million, up 88% year over year.

 

In other words, the widely cited $194 million consensus sits very close to the company's own core revenue guidance, and actual core revenue of $209.9 million exceeded it. Which measure you use to judge whether this report met the bar produces the opposite conclusion.

 

To be explicit, this article cannot fully reconstruct the reconciliation between the two measures from public reporting, and readers should verify against the company's official financial statements. But the discrepancy itself demonstrates that when assessing newly listed companies, confirming what a figure actually measures matters more than comparing its size.

 

 

Setting the measurement dispute aside, the segment data is clear. Investing.com's summary shows core cloud and other services revenue jumping 287% year over year to $127.7 million. That line grew far faster than the total, indicating the revenue mix is shifting toward services, which typically carries higher margins and more recurring characteristics.

 

 

 

This is the highest-information comparison in the report. Core gross margin was 40.6%, up roughly 940 basis points from 29.6% a year earlier, but down roughly 590 basis points from 46.5% in the first quarter.

 

A further comparison is often overlooked: the company's own second quarter core gross margin guidance was 36% to 38%. The actual 40.6% came in roughly 2.6 percentage points above the top of that range. Measured against the company's own expectations, margin beat.

 

Segment data is equally positive. Core cloud and other services gross margin was 41.8%, up 1,600 basis points year over year, while core hardware gross margin was 38.8%, up 510 basis points. Core operating margin of negative 16% improved sharply from negative 42% a year earlier.

 

 

Management's explanation is specific and verifiable. Per the earnings call transcript published via Yahoo Finance, the company said it is meeting some of the overwhelming demand for its fast inference service by temporarily renting some of its own systems back from cloud customers and making them available through the Cerebras cloud, which reduces gross margin in the short term because of the higher cost of that rented capacity, and that absent those higher private cloud costs core gross margin would have been approximately 500 basis points higher and more similar to the prior quarter.

 

The same transcript shows management guiding third quarter core gross margin to 38% to 40% while stating that the third quarter marks the low point for this year's gross margin.

 

The nature of that explanation matters: the margin decline reflects a premium paid to serve demand earlier rather than eroding pricing power. CNBC's report quoted CEO Feldman saying gross margins are in a good spot and growing because fast inference is priced at a premium, adding that Cerebras was able to increase the AI output of its systems.

 

For investors following both AI hardware and crypto, the competition between the two for high-end chips and power capacity is worth tracking. On venues offering crypto trading alongside stock-linked products, such as MEXC, relative performance between the two often signals shifts in capital preference early.

 

 

 

This is the most counterintuitive part of the episode. Reuters reported that Cerebras raised its annual revenue and gross margin forecasts and said it was not experiencing supply shortages as severe as other chip designers, expecting 2026 adjusted revenue of $880 million to $890 million against a previous forecast of $855 million to $865 million, and raising the annual adjusted gross margin projection to 41% to 43% from 38% to 41%, while management said it plans to more than triple revenue in 2027.

 

The call transcript adds that third quarter core revenue guidance is $214 million to $216 million with core operating margin guided to negative 25% to negative 23%, and full-year core operating margin raised to negative 19% to negative 17% from negative 28% to negative 32%. The operating margin revision is particularly large.

 

The order book supports the growth story too. Remaining performance obligations reached $25.4 billion as of June 30, roughly 29 times the $880 million to $890 million full-year revenue guide. The balance sheet is well capitalised, with $8.6 billion in cash, cash equivalents, restricted cash and marketable securities plus an unused $850 million revolving credit facility.

 

Against that dataset, the stock still fell roughly 16%. That says the market is not pricing the absolute quality of the data but the gap between the data and what was already embedded in the price.

 

 

This is the essential context. GuruFocus's summary notes the stock's price-to-sales ratio stands at an elevated 291.5 times, well above its historical median of approximately 238.6 times and the semiconductor industry average, signalling high market expectations for future revenue growth, while traditional earnings-based valuation metrics such as price-to-earnings are not applicable because the company is unprofitable and cash-flow negative.

 

A simple illustration at that multiple: at 291.5 times sales, each 1% shortfall in revenue corresponds to roughly a 1% loss of market value assuming the multiple holds. Reported revenue of $180.11 million against a $194.23 million estimate is a shortfall of roughly 7.3%. That is a rough demonstration from public data rather than an explanatory model of the price reaction, but it illustrates how any revenue deviation gets amplified at extreme sales multiples.

 

Price history offers another reference. Per FX Leaders' analysis, the company's 2026 IPO was the largest semiconductor offering on record, raising $6.4 billion at 20 times oversubscription against an initial price range of $125 to $135, with the stock closing at $350 within its first days of trading, an unprecedented 2.6 times move. The stock traded near $234.85 ahead of earnings, already well off that $350 peak.

 

The same analysis noted Wall Street maintaining a Strong Buy consensus with an average price target of $297.56. A consensus target above the current price coexisting with a decline on raised guidance suggests actual positioning expectations ran higher than the published ratings implied.

 

 

The first risk is customer concentration. Reporting indicates revenue is concentrated in OpenAI and AWS, customers whose purchasing decisions are highly correlated, so a change of cadence at either would materially affect the revenue curve. The call transcript also noted that a meaningful portion of data centre fit-out capital expenditure is reimbursed as pass-through cost from the largest customer, underscoring that dependence.

 

The second is the margin recovery path. Management said the third quarter is this year's low point, with third quarter guidance at 38% to 40% against a full-year target of 41% to 43%. That implies the fourth quarter must run well above the midpoint of the annual range. If the third quarter prints at the low end, achieving the full-year target becomes materially harder.

 

The third is valuation itself. At roughly 291.5 times sales, any revenue shortfall is amplified, and with the company unprofitable and cash-flow negative there is no earnings metric available as valuation support.

 

The fourth is the communication cost of inconsistent revenue measures. With a material gap between reported and core revenue, the same report can be simultaneously described as a beat and a miss depending on which measure a given source uses, which itself amplifies volatility.

 

On scenarios, the base case is third quarter core revenue landing within the $214 million to $216 million guide with margin bottoming, shifting attention to the feasibility of tripling revenue in 2027. A second is the rented-capacity cost fading faster than expected and margin returning above 46%, which would support the multiple. A third is another revenue print below consensus, where, with the multiple near 300 times sales and no earnings support, the drawdown could exceed this one.

 

Status distinctions: financial data and guidance come from the company's official release and public call records. After-hours prices and decline percentages come from third-party sources, change constantly and differ between providers. The price-to-sales ratio and price targets are third-party calculations and judgements. The reconciliation between the two revenue measures cannot be fully reconstructed here and should be verified against the company's official statements.

 

 

What deserves recording about this decline is that a report which beat comprehensively against the company's own guidance framework was judged a failure by the market. Core gross margin of 40.6% exceeded the top of the 36% to 38% guide, core operating margin of negative 16% beat guidance of negative 30% to negative 32%, and both full-year revenue and gross margin guidance were raised, with the operating margin revision exceeding ten percentage points. These are verifiable facts. The stock still fell roughly 16%, and the reason lies not in the data but in the near-zero margin for error implied by roughly 291.5 times sales.

 

Three misreadings look likely. The first is treating the revenue shortfall as evidence that growth is stalling. The gap between $180.11 million and $194.23 million is real, but the core revenue the company emphasised was $209.9 million, above its own prior core revenue guidance of roughly $194 million. Two measures produce opposite conclusions, and absent a full reconciliation neither conclusion is secure. The second is reading the sequential margin decline as eroding pricing power. Management explained it as temporarily renting back its own systems from customers to serve inference demand, stated that margin would have been roughly 500 basis points higher without that cost, and identified the third quarter as this year's low point. That is a deliberate capacity choice rather than passive price pressure, and the two imply very different futures. The third is underestimating how the multiple amplifies the reaction. With a sales multiple near 300 times and no earnings metric available as a reference, a roughly 7% revenue deviation translated into roughly a 16% price move, and that relationship is itself the risk warning.

 

Three verifiable metrics deserve tracking. First, where third quarter core gross margin lands relative to the 38% to 40% guide. Management has committed to this being the annual low, so a print at or below the bottom would require an unusually strong fourth quarter rebound to reach the 41% to 43% full-year target, and step-change improvements of that kind carry low credibility. Second, the pace at which the rented-capacity cost fades, which management quantified at roughly 500 basis points and which can be verified directly in subsequent margin prints. Third, whether revenue reporting becomes consistent, since continuing to present two materially different figures would keep producing mismatches between consensus and disclosure that trigger similar volatility.

 

For cross-asset investors, the lesson is that the valuation multiple determines the tolerance radius of any given report. On an asset trading near 300 times sales, a 7% revenue deviation is enough to produce a 16% price reaction, while on an asset trading at single-digit sales multiples the same deviation is typically absorbed entirely. The same principle holds in crypto, where newly launched tokens with very high fully diluted valuations are acutely sensitive to any data point that disappoints, while mature assets whose valuations have been digested react far less. When evaluating names like this, establish what growth assumption the current multiple implies before judging how far the actual data sits from that assumption. That sequence gets closer to how prices form than comparing data quality alone. From a risk management standpoint, in earnings windows for high-multiple assets, controlling position size is worth far more than calling direction, because valuation has already compressed the tolerance band to a narrow range.

 

This analysis rests on the company's official release, public earnings call records and credible reporting available now. Subsequent quarterly data, the margin recovery path and customer composition could each change the conclusion, and no single scenario should be treated as a fixed expectation.

 

 

 

The immediate trigger was a revenue shortfall. Reported revenue of $180.11 million came in below the $194.23 million LSEG consensus, a gap of roughly 7.3%. The deeper reason is valuation tolerance. The stock trades near 291.5 times sales against a historical median of roughly 238.6 times, and with the company unprofitable and cash-flow negative there is no earnings metric to anchor valuation. In that structure any revenue deviation is significantly amplified.

 

 

It depends on the measure. Reported revenue of $180.11 million fell below the $194.23 million consensus. But the core revenue the company highlighted on its call was $209.9 million, up 103% year over year, against its own prior second quarter core revenue guidance of approximately $194 million, meaning it exceeded that guide. The full reconciliation between the two figures should be verified against the company's official financial statements.

 

 

It moved in two directions. Core gross margin of 40.6% rose roughly 940 basis points from 29.6% a year earlier and exceeded the company's own 36% to 38% second quarter guidance, but fell from 46.5% in the first quarter. Management attributed the sequential decline to temporarily renting back its own systems from cloud customers to meet inference demand, said margin would have been roughly 500 basis points higher without that cost, and identified the third quarter as this year's low point.

 

 

The revisions were broad. Full-year core revenue guidance rose to $880 million to $890 million from $855 million to $865 million; full-year core gross margin guidance rose to 41% to 43% from 38% to 41%; and full-year core operating margin rose to negative 19% to negative 17% from negative 28% to negative 32%. Third quarter guidance calls for core revenue of $214 million to $216 million and core gross margin of 38% to 40%. Management also said it plans to more than triple revenue in 2027.

 

 

It indicates high revenue visibility. Remaining performance obligations stood at $25.4 billion as of June 30, roughly 29 times the $880 million to $890 million full-year revenue guide. That says the order book far exceeds current delivery capacity, making capacity rather than demand the binding constraint. Recognition still depends on actual delivery, so remaining performance obligations are not equivalent to recognised revenue.

 

 

Liquidity is ample. At quarter end the company held $8.6 billion in cash, cash equivalents, restricted cash and marketable securities plus an unused $850 million revolving credit facility. It went public in May priced at $185, raising $6.4 billion in the largest semiconductor IPO on record at 20 times oversubscription. The company remains unprofitable, however, with core operating margin at negative 16%.

 

 

First, customer concentration, with reporting indicating revenue concentrated in OpenAI and AWS and the call noting that a meaningful portion of data centre fit-out capital expenditure is reimbursed as pass-through cost from the largest customer. Second, the margin recovery path, with third quarter guidance at 38% to 40% against a 41% to 43% full-year target implying a strong fourth quarter. Third, valuation itself, where roughly 291.5 times sales amplifies any deviation. Fourth, the communication cost of inconsistent revenue measures.

 

 

Three verifiable metrics. First, where third quarter core gross margin lands relative to the 38% to 40% guide, which management has committed to as this year's low point. Second, the pace at which the rented-capacity cost fades, quantified by management at roughly 500 basis points and directly verifiable in subsequent margin prints. Third, whether revenue reporting becomes consistent, since two materially different figures continuing to coexist would keep producing volatility-inducing mismatches.

 

 

This article is provided for informational and research purposes only and does not constitute investment advice, financial advice, legal advice, tax advice or any recommendation to transact. The financial figures and guidance referenced here come from the company's official release and public earnings call records; all figures covering the remainder of 2026 and beyond are company guidance rather than confirmed results and actual outcomes may differ materially; the reported revenue and core revenue figures cited here are different measures whose full reconciliation should be verified against the company's official financial statements; after-hours prices, decline percentages, price-to-sales ratios, price targets and consensus estimates come from third-party sources and calculations that change constantly and differ between providers; and every calculation from public data in this article is a simple arithmetic illustration rather than an explanatory model of price behaviour. Prices of crypto assets, equities and other related financial instruments can move sharply over short periods, and shares of highly valued, not-yet-profitable companies are especially volatile around earnings windows, with the possibility of total loss of principal. Historical performance, technical indicators and on-chain data cannot guarantee future outcomes. Readers should conduct their own independent research, verify official information directly, and evaluate any decision against their own financial circumstances, investment objectives, experience and risk tolerance, consulting a qualified professional adviser where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from use of or reliance on the information in this article.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise:

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)