Cerebras hands prefill to AMD GPUs to get 5x more tokens per megawatt

The production system puts AMD Instinct GPUs on prompt processing and Cerebras CS-3 systems on decode, first on Cerebras Inference Cloud and then in partner data centers.

By · Published

AI-generated illustration of server racks lining a data center aisle, with glowing blue and orange lines between them and an AMD label on the left.

The production system puts AMD Instinct GPUs on prompt processing and Cerebras CS-3 systems on decode, first on Cerebras Inference Cloud and then in partner data centers.