Nvidia rival Cerebras hits $95b market cap in Nasdaq debut

Cerebras Systems jumped 68% in its Nasdaq debut on May 14, ending at US$311.07 after pricing shares at US$185, higher than expected.

The AI chipmaker was valued at roughly US$95 billion.

The offering was the biggest US tech IPO since Uber in 2019 and could reach US$6.38 billion if underwriters buy 4.5 million more shares.

Cerebras reported 2025 revenue of US$510 million and net income of US$88 million, up from a US$481.6 million loss a year earlier.

The company first filed in September 2024 but withdrew after scrutiny of its reliance on G42, and its updated filing said G42 made up 24% of 2025 revenue while Mohamed bin Zayed University accounted for 62%.

🔗 Source: CNBC

🧠 Food for thought

Implications, context, and why it matters.

Cerebras posted profit on paper while its core business still lost money

Cerebras reported Generally Accepted Accounting Principles (GAAP) net income of about US$237.8 million in 2025. That figure came from a one-time, non-cash gain of about US$363 million tied to a forward contract liability, while the company logged an operating loss of about US$145.9 million 1.
Cerebras now leans on a multi-year Master Relationship Agreement with OpenAI for 750 megawatts (MW) of inference capacity, with an option for more, plus a tie-up with Amazon Web Services (AWS), Amazon’s cloud computing business, to widen revenue sources 2.
That focus likely helped demand lift the IPO from an expected US$115 to US$125 a share to a final price of US$185 a share 3.

The IPO and inference push add pressure in AI hardware

The IPO response suggests specialized chip designs beyond the graphics processing unit (GPU) model can win AI workloads, especially inference, which is the process of running trained AI models to generate results. The available sources do not back a broader claim that the market is breaking from near-total GPU dominance 4.
The claim that Nvidia acquired assets from inference startup Groq for US$20 billion is not backed by the provided sources and should not be treated as fact 5.
The AWS relationship suggests large cloud providers are adding alternative AI accelerators beside existing systems. That could widen the market for multi-accelerator cloud offerings aimed at business customers 2.