Cerebras Systems, the AI chipmaker, is leaning hard into its next-generation wafer-scale chip, the Wafer-Scale Engine 3 (WSE-3), as a cure for its struggling stock chart. The company's WSE-3 promises faster inference than traditional GPUs, with 125 petaflops of AI computing power and thousands of times greater memory bandwidth.
The WSE-3 measures 46,225 square millimeters and packs 4 trillion transistors, delivering 10x to 15x faster token generation for AI inference tasks. This architectural advantage could justify the switching costs for large cloud providers, according to Cerebras.
The company has landed partnerships with OpenAI and AWS to deploy its inference capabilities, but investors remain skeptical after a brutal post-debut selloff. The stock has cratered more than 50% from its highs, trading below the IPO price.
Cerebras raised approximately $5.55 billion through its IPO, earning an initial valuation near $56 billion. The company raised its 2026 revenue guidance to between $880 million and $890 million, a meaningful target for a company of its size.
However, the market will ultimately judge Cerebras on whether those collaborations translate into recurring, growing revenue rather than one-time deployments. For investors who bought at the IPO price, the stock would need to more than double from its August lows just to revisit its first-day highs.
The Challenge Ahead
Cerebras' hardware revenue has shifted unpredictably, and operational challenges after going public have drawn scrutiny from analysts watching profit margins. The company must now demonstrate its ability to execute on its vision and deliver on its revenue targets.
A New Chapter for Cerebras
Despite the challenges, Cerebras is poised to play a significant role in the growing AI market. Its WSE-3 chip offers a unique advantage in terms of memory bandwidth and inference performance, making it an attractive solution for large cloud providers.
As the company moves forward, investors will be watching closely to see whether Cerebras can overcome its current struggles and deliver on its promise of a new generation of AI computing hardware.
