A real-time fraud detection system uses Claude to analyze transaction patterns. The system must evaluate each transaction within 800ms end-to-end (including network round-trips). Input context per transaction is approximately 3,000 tokens including 30-day transaction history. The fraud team wants Claude to explain its reasoning for every flagged transaction. Current architecture uses Claude 3 Sonnet and achieves 1,400ms P95. Which architectural change most directly addresses the latency requirement without sacrificing explainability?

The option that switches to Claude 3 Haiku and implements streaming with partial-result rendering is architecturally superior because it addresses both requirements simultaneously. Switching to Haiku dramatically reduces time-to-first-token and generation speed for 3,000-token inputs — helping meet the 800ms requirement. Critically, streaming with partial-result rendering means the fraud flag decision, represented by the first substantive output token, can be emitted and acted on immediately while the explanation continues streaming to the analyst interface — preserving explainability without blocking on full response completion. The Haiku-only option switches to Haiku but ignores that the 800ms requirement includes full response generation for the explanation — streaming is the key architectural insight for decoupling the decision gate from explanation delivery. The 7-day history option reduces context window, potentially harming fraud detection accuracy by losing older pattern signals — a capability tradeoff that should be tested carefully, and the latency saving may be insufficient. The prompt-caching option helps with repeated system prompts but transaction data is unique per call — cache hit rates on the variable transaction history will be low, providing marginal latency improvement.