A real-time chatbot needs sub-second-class first tokens for simple FAQ answers. Which model choice is MOST appropriate as a default?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Route simple FAQs to Haiku; keep Sonnet/Opus for the hard stuff. That's classic speed/cost tiering.
Full explanation below image
Full Explanation
Haiku is optimized for low latency and cost on lighter tasks; production systems often use model routing. Always-Opus is expensive overkill for FAQs. Removing safety or inventing nonexistent models is wrong.