Cerebras CS-4: three wafers, 1000+ tokens/sec, 30x faster than GPUs
Cerebras unveiled the CS-4, a rack-scale system packing three WSE-3 Turbo wafers for up to 30x faster inference than GPU systems. Its modular Nexus architecture separates power and cooling from compute, enabling hyperscale deployment in hours and handling models beyond 10 trillion parameters at over 1,000 tokens per second.