Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Generative AI Leader practice bank has 2,191 more like it. Try 10 free: instant score, no account.
An enterprise is designing an automated customer support triage system that processes millions of incoming user requests each day. The system must quickly classify user intent and route inquiries in near real-time with minimal response latency and low inference costs, while maintaining adequate quality for routine text processing.
Which model tier should the enterprise select to best balance these operational and cost requirements?
Un-tuned base foundation models (such as raw pre-trained variants of open weights models) are foundational architectures trained solely on vast text corpora for raw sequence prediction, without conversation or instruction fine-tuning.
Flagship models, such as Gemini Pro, are large-parameter foundation models optimized for advanced reasoning, complex coding, multi-step problem solving, and nuanced domain-specific analysis.
Specialized generative models (such as dedicated text-to-video or text-to-audio generation engines) are built to synthesize rich multimedia assets, such as synthetic speech or video clips.
Lightweight model tiers, such as Gemini Flash-Lite and Gemini Flash, are streamlined foundation models designed specifically for high-speed, cost-sensitive, and high-frequency production workloads. They offer lower parameter footprints compared to flagship models while retaining strong general language understanding.
For high-frequency, relatively straightforward tasks like ticket triage and classification, deploying top-tier reasoning models introduces unnecessary cost and latency. A lightweight tier balances speed, affordability, and task accuracy perfectly.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated