Courseiva

CCDV-F Model Selection and Cost Management Practice Question

You are designing an AI agent that performs multi-step reasoning. The first step involves basic data extraction, while the second step requires complex logical deduction based on the extracted data. How should you select models to optimize for both performance and cost?

⚠ Common exam trap

Test-takers often default to using a single expensive model for an entire multi-step workflow, wasting budget on simple preliminary tasks.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Use Claude 3 Haiku for extraction and Claude 3.5 Sonnet for logical deduction.

Model routing is a sophisticated cost management technique where different tasks within a single workflow are assigned to the most appropriate model. Using a 'one-size-fits-all' approach often leads to waste. By chaining models—using a cheaper model for simple tasks and a more powerful one for complex tasks—developers can achieve high-quality results while minimizing the total token spend for the entire process.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • ✗

    Use Claude 3 Opus for both steps to ensure maximum consistency.

    Why it's wrong here

    Using Opus for the initial data extraction is an inefficient use of resources, as the task does not require its advanced reasoning capabilities. This approach significantly increases the cost of the agent without providing a meaningful improvement in extraction accuracy, as smaller models like Haiku are already highly proficient at simple structured data extraction tasks.

  • ✗

    Use Claude 3 Haiku for both steps to minimize total operational costs.

    Why it's wrong here

    While this is the cheapest option, Claude 3 Haiku may struggle with the 'complex logical deduction' required in the second step. This could lead to failures in the agent's logic, resulting in incorrect outputs. Cost savings are only beneficial if the model can successfully complete the task; otherwise, the lower cost results in a non-functional application.

  • ✓

    Use Claude 3 Haiku for extraction and Claude 3.5 Sonnet for logical deduction.

    Why this is correct

    This tiered approach, known as model routing or chaining, optimizes for both cost and intelligence. Haiku handles the high-volume, low-complexity extraction task cheaply and quickly, while the more expensive Sonnet is reserved for the difficult reasoning phase. This balance ensures the agent remains reliable while keeping the overall cost per execution much lower than using Sonnet alone.

  • ✗

    Use Claude 3.5 Sonnet for extraction and Claude 3 Haiku for logical deduction.

    Why it's wrong here

    This configuration is counter-intuitive and inefficient. It uses a more expensive model for the simple task and a less capable model for the difficult task. This would likely result in higher costs for the first step and a high probability of failure in the second step, representing a poor architectural choice for both performance and cost management.

About these practice questions

This CCDV-F question is part of Courseiva's 257-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

JA

Written and reviewed by Johnson Ajibi, MSc IT Security

Senior Network & Security Engineer · founder of Courseiva

Last reviewed September 2026 · checked against the official Anthropic exam blueprint

This CCDV-F practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCDV-F exam.