Cerebras Systems

Cerebras Systems, founded in 2015, designs and builds a new class of artificial-intelligence supercomputers built around a wafer-scale architecture. Its central product is an AI chip that is 56 times larger than a conventional GPU, delivering the compute power of dozens of GPUs on a single piece of silicon while maintaining programming simplicity.

The company operates globally and works across the AI computing, machine learning, and healthcare sectors. Its engineering work spans computer architecture, deep learning, AI computing, and hardware engineering, reflecting a focus on pushing the physical limits of chip design for large-scale AI workloads.

Cerebras positions itself at the frontier of what is technically achievable in AI hardware. The wafer-scale approach departs from the multi-chip GPU cluster model that currently dominates AI training, aiming to reduce the complexity and overhead associated with distributing work across many smaller processors.