A senior developer is optimizing their token costs. They discover that 60% of their API cost comes from…

Question

A senior developer is optimizing their token costs. They discover that 60% of their API cost comes from input tokens (most content is repeated context). Only 15% comes from output tokens, and 25% from thinking tokens. What optimization provides the biggest cost reduction?

Accepted Answer

D. Implement prompt caching to save on the 60% input cost — cache reads cost only 10% of base input price

Answer

A. Reduce output length with max_tokens to save on the 15% output cost

Answer

B. Reduce thinking budget to save on the 25% thinking cost

Answer

C. Switch to a smaller model that has lower per-token costs across all categories

Question 263