$$\text{price per token} \gtrsim \frac{C_{\text{fixed}} + C_{\text{energy}}}{U \cdot T_{\max}} + m$$

Equation 1 · AI Industry

How AI Inference Economics Actually Work

This mathematical expression combines the displayed quantities; its precise role follows from the surrounding article text.

This is the reason batching and caching matter for price rather than only for latency: a serving system that keeps a GPU at 80% average utilization across a day divides its fixed hourly cost across roughly twice as many tokens as one running at 40%, and can profitably charge roughly half as much per token for the same margin. Independent benchmarking gives some visibility into what utilization current hardware and software combinations can actually achieve under realistic load. MLCommons’ MLPerf Inference benchmark suite tests submitted systems under both a latency-bounded “server” scenario and an unconstrained “offline” scenario designed to maximize throughput through batching, and recent…

Meanings in this article

• $C_{\text{fixed}}$: the amortized hardware and facility cost per unit time.
• $C_{\text{energy}}$: the energy cost per unit time.
• $T_{\max}$: the chip’s maximum achievable token throughput under ideal batching.
• $m$: the margin term.

Equation guide → · Article →