How It Works
Perplexity processes user queries by first searching the web in real-time, then using large language models to synthesize the results into a coherent, cited answer. Every claim in the response is linked to its original source with numbered inline citations. Users can choose between Quick Search for fast answers and Pro Search for deeper, multi-step research. Focus modes allow filtering results by academic papers, YouTube videos, Reddit discussions, or specific domains. The platform also supports file uploads for document analysis and Collections for organizing research.