AI Inference Cost
The actual compute expenditure incurred during the inference phase of an AI model—that is, when the model responds to user requests. It encompasses GPU cluster usage, electricity consumption, and hardware depreciation, and constitutes a central cost component in the commercialization of AI products.
Timeline (last 90 days)
The cost of reaching a fixed level of AI performance fell from $20 per million tokens in late 2022 to $0.07 about eighteen months later, a price drop of more than 280 times.
Sam Altman once promised that the cost of AI intelligence would drop tenfold every year.
All Facts (3)
The cost of reaching a fixed level of AI performance fell from $20 per million tokens in late 2022 to $0.07 about eighteen months later, a price drop of more than 280 times.
50%UnverifiedSam Altman once promised that the cost of AI intelligence would drop tenfold every year.
50%Unverified当AI工具普及后，纯编码的市场价格会持续下降
70%