Maximize AI Factory Energy Efficiency Through Full-Stack Inference and Training Optimizations
Power can reach 40% of AI factory operating expenses, making performance per watt a critical efficiency metric.
“Power can account for 40% of the operating expenses (OpEx) to run an AI factory.”
NVIDIA outlines full-stack inference and training optimizations to maximize AI factory energy efficiency, noting power can be 40% of operating expenses. As most sites face fixed power caps, performance per watt directly translates to token costs, making it a key competitive lever for AI infrastructure operators.