Q7 — AWS AIF-C01 Ch.2

Question 7 of 100 | ← Chapter 2

A company is using few-shot prompting with a foundation model hosted on Amazon Bedrock. The model currently uses 10 examples in its prompt. The model is invoked once per day and performs well. The company wants to reduce monthly costs. Which solution satisfies these requirements?

Correct Answer: B. Reduce the number of tokens in the prompt.

Explanation

Costs for foundation models hosted on Amazon Bedrock are directly tied to the number of tokens processed in prompts and responses. Since the model currently uses 10 examples per prompt, is invoked only once daily, and performs well, reducing the number of tokens in the prompt is the most effective way to lower monthly costs without compromising performance. Fine-tuning may improve task-specific accuracy but does not inherently reduce inference costs; increasing token count would raise costs; and provisioned throughput improves latency and scalability—not cost efficiency. Therefore, reducing the number of tokens in the prompt is the optimal solution to meet the cost-reduction requirement.