Key idea: Multi-agent means splitting work across specialized model calls. It can help when each agent brings its own data, but every handoff costs time and money.
- Send the same idea as Single agent, then as Team. Compare the score, time, and cost.
- Switch to Team + critic. Does the review round raise the score enough to justify the extra calls?
- Turn off Specialists in parallel. Same work and cost, but watch the time go up.
- Try Single agent on the Smart model against Team on the Fast model. Which wins, and at what cost?
- Send the same idea twice with the same settings. Scores move a point or two: judge one run with care.
Try a prompt