A team wants to compare the cost-effectiveness of two LLMs they are evaluating for a summarization task. They have run both models on the same set of documents and recorded the token usage. They need to calculate the total cost for each model. What information do they need in addition to token usage?

The number of GPUs used during training

The number of users who accessed the model

The model's accuracy on a benchmark

The price per 1,000 tokens for each model

A team wants to set up a dashboard that shows daily token consumption and cost per model serving endpoint. They want to use Databricks-native tools. What should they use?

Use the Databricks CLI to manually check endpoint status.

Export cluster logs to a third-party monitoring tool.

Create a SQL query against the system tables and build a dashboard in Databricks SQL.

Use the MLflow UI to view serving endpoint metrics.

A team wants to reduce LLM costs for a chatbot that frequently receives the same user questions. Which strategy is most effective for lowering token consumption?

Increasing the max tokens parameter

A data engineer needs to analyze token consumption and associated costs for LLM endpoints across the organization. Which Databricks tool is specifically designed for this purpose?

System tables for billing and usage

A data science team uses Databricks Model Serving to host a fine-tuned LLM. They notice that the monthly bill is higher than expected. They need to identify which specific models and endpoints are driving the token consumption. Which Databricks tool should they use?

The System Tables for billing and usage, querying the usage table for token counts per endpoint.

The Databricks SQL warehouse query history.

The MLflow experiment tracking page for the model.

The cluster event logs for the serving endpoint.

DatabricksCertified Generative AI Engineer Associate

Domain 6Objective 4

Cost Control and Usage Tracking GENERATIVE-AI-ENGINEER-ASSOCIATE Practice Questions (Page 1)

Part of the Section 6: Evaluation and Monitoring domain, which makes up ~17% of our current practice bank. Databricks does not publish an official question count, but from its 90-minute exam (~35–60 total, ~6–10 in this domain), expect 1–2 from this objective — we provide 21 practice questions to prepare you well beyond it. (estimate)

21questions here

5free pages

3concepts

Questions 1–5

1application · easy
A team wants to compare the cost-effectiveness of two LLMs they are evaluating for a summarization task. They have run both models on the same set of documents and recorded the token usage. They need to calculate the total cost for each model. What information do they need in addition to token usage?
Select an answer first
2application · medium
A team wants to set up a dashboard that shows daily token consumption and cost per model serving endpoint. They want to use Databricks-native tools. What should they use?
Select an answer first
3foundation · easy
A team wants to reduce LLM costs for a chatbot that frequently receives the same user questions. Which strategy is most effective for lowering token consumption?
Select an answer first
4foundation · easy
A data engineer needs to analyze token consumption and associated costs for LLM endpoints across the organization. Which Databricks tool is specifically designed for this purpose?
Select an answer first
5application · medium
A data science team uses Databricks Model Serving to host a fine-tuned LLM. They notice that the monthly bill is higher than expected. They need to identify which specific models and endpoints are driving the token consumption. Which Databricks tool should they use?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to GENERATIVE-AI-ENGINEER-ASSOCIATE

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Databricks. “GENERATIVE-AI-ENGINEER-ASSOCIATE” is a trademark of its owner, used for identification only.