Your multi-agent architecture includes (1) a planner that breaks ambiguous goals into steps, (2) a high-volume classifier that routes tickets into ten categories, and (3) a code-generation specialist for infrastructure-as-code snippets. How should you match tasks to model capacities?
Select an answer to reveal the explanation.
Short Explanation
Correct answer: B. Architects must match task demands to model family capacities. Planners and complex code generation benefit from stronger reasoning models; high-volume routing often fits smaller, cheaper, lower-latency models. One-size-fits-all large models (A) inflate cost. Embeddings alone (C) cannot plan or generate code. Weak planners (D) collapse orchestration quality. Document model choices per agent role.
Full Explanation
“Match task demands to model family capacities” is an explicit architect skill. Heterogeneous multi-agent systems intentionally use different deployments: frontier-class models for ambiguous planning and hard generation, lightweight models for deterministic-ish classification or extraction, and specialized models where available. Consider context window needs, tool-calling reliability, latency SLOs, and unit economics. Option A ignores cost and latency. Option C misuses embedding models. Option D inverts priority—planning errors cascade. Include fallback models and evaluation criteria so model swaps remain safe.