Perplexity

Perplexity AI · December 2022

activeCloseddecoder onlytext

Why It Matters

Pioneered the 'answer engine' category — combining real-time web search with LLM synthesis to provide cited, up-to-date answers. Became the first serious challenger to Google Search.

Description

An 'answer engine' that combines real-time web search with large language model reasoning to provide direct, cited answers instead of a list of links. Unlike traditional search engines, Perplexity reads and synthesizes information from multiple sources, then presents a coherent answer with inline citations so users can verify the information. Supports follow-up questions for iterative research.

Notable Milestones

  • First mainstream AI-powered search alternative to Google
  • Cited answers with real-time web data

Key Innovations

Tool Use
Tool UseAbility to call external tools, APIs, and functions — enabling web browsing, code execution, and real-world actions.
Reasoning
ReasoningStructured step-by-step problem solving, often using chain-of-thought or tree-of-thought approaches.

Related Research (1)

RAGReasoning
2020 · Facebook AI Research / UCL

Combined a neural retriever (that searches a knowledge base) with a sequence-to-sequence generator, allowing the model to 'look up' relevant documents…