Plurality sends your question to many models in a structured round, then returns one reconciled answer — with the agreements, the dissents, and the citations laid out next to it.
Your question is parsed, the panel is fanned out, and each model gets the same prompt — in parallel, so none of them can anchor on another.
What you see below is roughly what shows up on screen during a session — six independent reads on the same question, returned within seconds of each other.
Plurality doesn't average the answers — it shows you exactly where the panel agreed, where it split, and what it cited. You get one document you can read in a minute or audit in a day.
The mode is the rules of engagement — how many rounds, what each model is asked to do, how the answer is reconciled. Seven options. One you'll use most.
The panel is who joins the deliberation. A preset is enough for most questions; build your own when you need a specific mix.
You can do meaningful work on the free tier. Pro is for daily use; Team is for shared workspaces and audited workflows.
Most users get a first deliberation back in under ten seconds. You'll know within one whether this is for you.