Cerebras Systems unveils CS-4 processor and signs deal with OpenAI for 750 megawatts of capacity
Cerebras Systems unveiled a new generation of its wafer-scale processor, the CS-4, and signed a multi-year agreement with OpenAI to deploy 750 MW of compute capacity in 2026–2028.
In August 2026, Cerebras Systems unveiled the CS-4, a new generation of its wafer-scale processors for AI computation. At the same time, it signed a multi-year agreement with OpenAI to deploy Cerebras systems with a total capacity of 750 megawatts between 2026 and 2028.
According to Cerebras Systems, as of August 2026 the company had more than 600 megawatts of data center capacity either already in operation or contractually secured for delivery by the end of 2027. The company also plans to launch its first European data center capacity still in 2026 and expand it to 200 megawatts by the end of 2027. According to the company's own statement, it intends to increase its manufacturing capacity more than tenfold during 2026.
Cerebras Systems previously went public via an IPO in May 2026, which according to the source raised $5.5 billion. The company has long built on a wafer-scale processor architecture, where the chip is manufactured directly on the entire silicon wafer instead of being cut into individual dies, which it presents as an alternative to traditional AI chips.
The available source is primarily an invitation to the TechCrunch Disrupt 2026 conference, where Cerebras Systems CEO Andrew Feldman is set to speak on the topic of scaling AI compute. For details on the technical specifications of the CS-4, see the source article.
Why it matters
For companies planning their own AI infrastructure or purchasing cloud compute capacity, the relevant information is the growing competition among chip suppliers and the fact that large customers like OpenAI are entering multi-year contracts for hundreds of megawatts of capacity – this suggests the direction in which AI compute supply, and potentially pricing, will develop. For the general public and individual users, the news has no immediate practical impact.
Relevant practical impact
What this means
For a business
The deal shows that OpenAI is diversifying its compute capacity suppliers beyond traditional GPU vendors, which may affect the availability and pricing of AI compute capacity on the market in the coming years.
StrategyCheck the original
Event sources
only one source so far · 1 publisher, 1 independent. We count feeds from the same owner only once.