Cerebras Artificial Intelligence Chipmaker Debuts on Stock Market

Cerebras Systems surged 68% on its Nasdaq debut, closing at $311.07 after selling shares at $185, valuing the chipmaker at about $95 billion.

Objective Facts

Cerebras Systems soared 68% in its Nasdaq debut on Thursday, closing at $311.07 after selling shares at $185, well above the company's expected range, valuing the chipmaker at about $95 billion. The company sold 30 million shares in its offering late Wednesday, raising $5.55 billion, the largest initial public offering for a U.S. tech company since Uber's debut in 2019. Revenue at Cerebras jumped 76% last year to $510 million, and the company generated net income of $88 million, swinging from a loss of $481.6 million a year earlier. The company announced a cloud deal with OpenAI in January worth more than $20 billion and partnered with Amazon Web Services to deploy its chips in data centers. Customer concentration remains an issue: the Mohamed bin Zayed University of Artificial Intelligence in the UAE accounted for 62% of revenue last year, down from G42's 85% concentration in 2024.

Left-Leaning Perspective

CNBC's Jim Cramer said Cerebras has a compelling AI story and promising technology, but warned the chipmaker's post-IPO valuation has become difficult to justify. Cramer said investors should avoid chasing shares after the stock's explosive debut, arguing the market is already pricing in years of massive growth. "While there might be a situation in the future where I can recommend Cerebras, I just can't even come close to justifying the valuation up here given how much it's already run right out of the gate," the "Mad Money" host said. "For now, I say keep your bat on your shoulder and hope the stock gives you a giant pullback. Because at these levels, it's too rich for me." An analysis from Benzinga argued the question is not whether AI infrastructure is real, but whether this valuation on this revenue base leaves room to generate a return. The GAAP net income of $237.8 million came almost entirely from a one-time non-cash gain of $363.3 million on a forward contract liability tied to G42. Stripping that out, the operating business posted a $145.9 million operating loss. As a result, investors buying CBRS are paying a growth premium for a business that has not yet demonstrated it can grow profitably. The Motley Fool's Matt Frankel noted that given the current IPO range and Cerebras' $510 million in 2025 revenue, the stock is likely to go public at about 95 times sales, noting that there's a lot of future growth priced in. Left-leaning coverage emphasized that Cerebras' revenue is heavily concentrated—about 86% of its 2025 revenue came from just two UAE-linked customers. As of market close on Thursday, the stock trades at more than 130 times sales, well above the multiples on much larger and more profitable chip companies like Nvidia.

Right-Leaning Perspective

Constellation Research analyst Holger Mueller argued that Cerebras provides a viable alternative to Nvidia for inference, noting that strong competitors will help reduce the cost of AI for enterprises and keep innovation flywheel spinning. Mueller pointed to Elon Musk's SpaceX, OpenAI and Anthropic as expected to go public later in the year in even bigger potential IPOs. An analysis from Robert Castellano's Substack noted that what changed is demand validation—a multi-year agreement with OpenAI and emerging deployment pathways through hyperscalers shift Cerebras from a single-customer system vendor to a candidate for infrastructure-level adoption. The IPO represents the first real attempt to price inference economics as a distinct segment of AI compute, separate from Nvidia's training dominance. The Motley Fool argued that Cerebras, with a massive backlog relative to current revenue, deserves a premium valuation and already has the backlog in place to grow revenue tenfold within a few years. Notably, none of Nvidia, Broadcom, and AMD are expected to grow their top lines as quickly as Cerebras. Right-leaning and bullish coverage highlighted strategic partnerships with OpenAI and Amazon as significantly increasing credibility. OpenAI already launched an AI model running on Cerebras hardware earlier in the year, validating real-world deployment potential. Bullish analysis from TradingView's Zacks noted that the Cerebras IPO highlights that investors are searching aggressively for the "next wave" of AI infrastructure winners. While Cerebras is intriguing, innovative, and potentially disruptive, it remains in a very early stage compared with Nvidia's ecosystem. For now, Nvidia looks like the safer AI infrastructure investment, but increased AI chip competition could matter more in the long term.

Deep Dive

Cerebras' most formidable competitor in hardware is Nvidia, the world's most valuable company. Cerebras claims speed and price advantages over graphics processing units from Nvidia due to architecture differences. In December, Nvidia paid $20 billion for assets from startup Groq, whose chips more closely resemble Cerebras, and later announced plans for Groq-based products. Independent benchmarks show Cerebras has a generational advantage over GPUs in inference scenarios, but the key word is "inference." Cerebras' strength lies in latency-sensitive inference workloads, not in challenging Nvidia's capabilities for large model training and general-purpose computing. Critical observers correctly identify that Cerebras' profitability metrics rely on one-time accounting gains, with real operating losses continuing. Bullish observers fairly note the company's legitimate backlog—$24.6 billion in contractually committed revenue—which would justify substantial growth if executed. Both sides acknowledge customer concentration, though left-leaning analysis treats it as a fatal flaw while right-leaning analysis views it as a problem mitigated by OpenAI and AWS deals providing distribution channels that traditional hardware vendors lack. The core disagreement centers on whether inference represents a large enough market segment to sustain 95x price-to-sales multiples and whether Cerebras can execute at scale in manufacturing and data center operations it has never attempted before. Key watch points include the share unlock schedule beginning August 2026, when approximately 84 million shares become eligible for sale by August 19, with the full 171.1 million shares unlocked by end of October 2026. That rolling supply overhang arrives just before Q3 earnings and will ultimately reveal whether insiders believe this valuation holds, or whether they use it as an exit. Even bullish Cerebras analysts acknowledge that Nvidia's latest Blackwell systems are narrowing performance gaps in inference workloads.

OBJ SPEAKING

Create StoryTimelinesVoter ToolsRegional AnalysisPolicy GuideAll StoriesCommunity PicksUSWorldPoliticsBusinessHealthEntertainmentTechnologyAbout

Cerebras Artificial Intelligence Chipmaker Debuts on Stock Market

Cerebras Systems surged 68% on its Nasdaq debut, closing at $311.07 after selling shares at $185, valuing the chipmaker at about $95 billion.

May 14, 2026· Updated May 15, 2026
What's Going On

Cerebras Systems soared 68% in its Nasdaq debut on Thursday, closing at $311.07 after selling shares at $185, well above the company's expected range, valuing the chipmaker at about $95 billion. The company sold 30 million shares in its offering late Wednesday, raising $5.55 billion, the largest initial public offering for a U.S. tech company since Uber's debut in 2019. Revenue at Cerebras jumped 76% last year to $510 million, and the company generated net income of $88 million, swinging from a loss of $481.6 million a year earlier. The company announced a cloud deal with OpenAI in January worth more than $20 billion and partnered with Amazon Web Services to deploy its chips in data centers. Customer concentration remains an issue: the Mohamed bin Zayed University of Artificial Intelligence in the UAE accounted for 62% of revenue last year, down from G42's 85% concentration in 2024.

Left says: Left-leaning analyst Jim Cramer acknowledged Cerebras' compelling story but warned the post-IPO valuation is unjustifiable and investors should avoid chasing shares, arguing the market is already pricing in years of massive growth.
Right says: Investors are aggressively searching for alternatives to NVIDIA in the accelerating AI compute race, and Cerebras is entering public markets with a differentiated architecture, strong industry partnerships, and bold claims about AI compute needs.
✓ Common Ground
Several analysts agree that the successful Cerebras IPO provided a bullish litmus test of investor demand for new AI listings.
Multiple observers note that Cerebras enters the public market with more institutional demand than supply, a condition that may support first-day performance above the offering price. Cerebras' positioning has changed materially since its 2024 withdrawal as the AI hardware market has rotated from training-cycle dominance.
Both skeptics and supporters acknowledge Independent Artificial Analysis benchmark tests show that the CS-3 outputs over 2500 tokens per user per second, compared to approximately 1000 tokens for NVIDIA's flagship DGX B200, with Cerebras having a generational advantage over GPUs in inference scenarios.
Across the spectrum, analysts recognize that Cerebras' backlog provides legitimate revenue visibility. The company's backlog climbed to $24.6 billion at the end of 2025, with potential to grow revenue tenfold within a few years if execution succeeds.
Objective Deep Dive

Cerebras' most formidable competitor in hardware is Nvidia, the world's most valuable company. Cerebras claims speed and price advantages over graphics processing units from Nvidia due to architecture differences. In December, Nvidia paid $20 billion for assets from startup Groq, whose chips more closely resemble Cerebras, and later announced plans for Groq-based products. Independent benchmarks show Cerebras has a generational advantage over GPUs in inference scenarios, but the key word is "inference." Cerebras' strength lies in latency-sensitive inference workloads, not in challenging Nvidia's capabilities for large model training and general-purpose computing.

Critical observers correctly identify that Cerebras' profitability metrics rely on one-time accounting gains, with real operating losses continuing. Bullish observers fairly note the company's legitimate backlog—$24.6 billion in contractually committed revenue—which would justify substantial growth if executed. Both sides acknowledge customer concentration, though left-leaning analysis treats it as a fatal flaw while right-leaning analysis views it as a problem mitigated by OpenAI and AWS deals providing distribution channels that traditional hardware vendors lack. The core disagreement centers on whether inference represents a large enough market segment to sustain 95x price-to-sales multiples and whether Cerebras can execute at scale in manufacturing and data center operations it has never attempted before.

Key watch points include the share unlock schedule beginning August 2026, when approximately 84 million shares become eligible for sale by August 19, with the full 171.1 million shares unlocked by end of October 2026. That rolling supply overhang arrives just before Q3 earnings and will ultimately reveal whether insiders believe this valuation holds, or whether they use it as an exit. Even bullish Cerebras analysts acknowledge that Nvidia's latest Blackwell systems are narrowing performance gaps in inference workloads.

◈ Tone Comparison

Left-leaning commentary from Jim Cramer expressed skepticism using phrases like "I just can't even come close to justifying the valuation up here" and "it's too rich for me," while right-leaning analysis framed investor interest positively, describing investors as "aggressively searching for alternatives" to Nvidia with "differentiated architecture" and "bold claims."