AI-102 Implement generative AI solutions Practice Question

Answer analysis

Option-by-option breakdown

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

• ✓

  Split the article into smaller sections and summarize each section separately.

  Why this is correct

  Splitting the article into smaller sections keeps each request within the model's context window, since a 10,000-token article exceeds it. Summarising each chunk separately, then optionally combining the partial summaries, avoids truncation and preserves coverage of the full content.

• ✗

  Increase the temperature parameter.

  Why it's wrong here

  Temperature controls sampling randomness, not token count, so the 10,000-token article still exceeds the context window and the request fails. It is tempting because temperature tuning shapes summary style, and it would be the right lever when output is too repetitive or too creative, not when input length is the constraint.

• ✗

  Use a model with a smaller context window.

  Why it's wrong here

  A smaller context window reduces capacity further, guaranteeing the 10,000-token article cannot fit at all. It is tempting because smaller models cost less and run faster, and choosing one would be correct for short prompts where latency and price matter more than handling long input.

• ✗

  Set max_tokens to a lower value.

  Why it's wrong here

  max_tokens caps only the generated completion length; the 10,000-token prompt still exceeds the context window and is rejected. It is tempting because it limits output size and cost, and it would be correct when you need shorter responses or want to control billing, not when the input itself is too long.

Go deeper

Related to this question