# Favley

> Favley is a multi-model AI consensus engine that queries 5–6 leading LLMs simultaneously, scores their agreement, and synthesizes a single audited verdict with a confidence score.

Favley exists because most AI tools give you one model's opinion presented as fact. Favley runs every query across multiple frontier models (including Claude, GPT-4o, Gemini, DeepSeek, Grok, and Perplexity Sonar), measures how much they agree, and returns a weighted synthesis along with a 0–100% consensus score — so disagreement between models becomes a visible signal of uncertainty or hallucination risk, not a hidden one.

## Product

- [Favley](https://favley.com): Homepage and consensus search — ask a question, get a multi-model audited answer with a confidence score.
- [Pro](https://favley.com/pro): Paid tier — full 6-model consensus, higher daily limits, clean PDF export.
- [AI Plagiarism Detector](https://favley.com/tools/ai-plagiarism-detector): Free tool that detects AI-generated text using a multi-model voting panel, with paragraph-level heatmap analysis for Pro users.
- [AI Tools](https://favley.com/ai-tools): Directory of Favley's free AI utilities.
- [Consensus Index](https://favley.com/consensus-index): Public, weekly-updated index tracking how AI model consensus on key topics shifts over time.

## Support

- [Contact](https://favley.com/contact): General contact and inquiries.
- [Support](https://favley.com/support): Help center and support ticket submission.
- [Privacy Policy](https://favley.com/privacy): How Favley handles user data.

## Optional

- [Terms of Service](https://favley.com/terms): Terms governing use of Favley.