3.1Standard
A conversational application needs quality and fast responses.
Compare candidate models under realistic demand.
Choose two appropriate practices.
× Incorrect
Measure quality, percentile latency, errors, concurrency, and cost using representative inputs and load.
Detailed explanation
CorrectPlan capacity using model size, output length, concurrency, and price
Operations require more than a benchmark score.
IncorrectUse benchmark rank as a guarantee of production speed
Workload and settings can differ.
CorrectMeasure quality, p50 and p95 latency, and errors with representative inputs
Tail latency and realistic quality matter.
IncorrectSelect the largest model without latency measurement
The requirements and budget may not be met.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock公式モデル選択・料金とAIF-C01 Domain 3.1を確認する。
Official references
What to confirm
品質、速度、容量、費用の比較軸を説明できる。
Key points
- p95
- Capacity
- Cost
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Capacity planning
Estimating traffic, concurrency, limits, latency, and cost to prepare adequate resources.
Percentile latency
A distribution value such as p50 or p95 that describes typical or tail response time.
Question IDAWS-AIF-162
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue