Features
Everything in the deliberation stack.
LLM Peers is a staged research system. Each feature below is a distinct part of getting from a difficult question to a decision-ready brief—clarifying intake, live web research, multi-model AI Council answers, anonymized peer ranking, chairman synthesis, and optional public sharing.
How the deliberation stack fits together
Unlike a single chat thread, LLM Peers separates generation, review, and synthesis. Optional intake sharpens goals and constraints. Web research can ground the board in current public context. The AI Council produces independent multi-model answers. Peer review anonymizes and ranks those answers against shared criteria. The chairman turns rankings into one decision brief. Public sharing makes a completed case readable for teammates without re-running the board.
That separation is intentional. Independent generation keeps alternative framings alive. Anonymized peer ranking reduces prestige bias. Synthesis closes the loop with a recommendation, ranked options, risks, assumptions, and a next action—so you leave with an artifact for judgment, not another prompt to write.
Explore each feature
Use the pages below for deeper detail on every stage. Together they describe the full product surface: from clarifying the decision to publishing a research case others can review.
Feature pages
- AI CouncilIndependent multi-model answers with competing recommendations, risks, and assumptions before peer review
- Peer review & rankingAnonymized comparative evaluation against shared criteria—substance over model brand preference
- Chairman decision briefFinal recommendation, ranked options, agreement and dissent, risks, and a practical next action
- Clarifying intakeOptional clarifying questions that sharpen goals, constraints, and success criteria before research
- Live web researchCurrent public sources as shared context so multi-model deliberation is not only training knowledge
- Public sharingPublish a research case link so others can read the question and decision brief without re-running
Do I need every feature on every run?
No. Intake and web research are optional depending on the brief. The core deliberation path—AI Council, peer review, and decision brief—is what turns a hard question into a ranked recommendation.
What makes this different from multi-tab AI chat?
Multi-tab chat still leaves you to compare and reconcile. LLM Peers runs independent generation, anonymized peer ranking, and chairman synthesis as one workflow with a decision-ready artifact.
Where should I start if I am new?
Start with How it works for the full path, then read AI Council, Peer review, and Decision brief. Use cases and Benefits show when the deliberation stack is worth running.