multiplication

The passage around this formula

That gap explains why distillation investment almost never fails its own break-even test once a product actually ships. Write the total cost of training and then serving a distilled student model as $TC_{\mathrm{distill}} = C_{\mathrm{distill}} + Q \cdot c_{\mathrm{student}}$. against the cost of simply continuing to serve the teacher directly for the same query volume,

Read this part in the article →