Sonar by Perplexity is an AI-powered search engine that allows users to select specific large language models (LLMs) to answer their queries. It provides summarized, cited answers, offering a customizable approach to information retrieval and research.
Claim this tool to publish updates, news and respond to users.
Sign in to claim ownership
Sign InSonar by Perplexity is an advanced AI-powered search engine that redefines how users interact with information online. Its core value proposition lies in offering a customizable and transparent research experience, where users are not limited to a single AI model but can select from a curated list of state-of-the-art large language models to generate answers. This approach combines the power of real-time web search with the reasoning capabilities of different LLMs, delivering concise, well-structured summaries that are directly cited to their original sources, thereby enhancing trust and verifiability.
Key features: The platform allows users to manually select an LLM, such as GPT-4o, Claude 3.5 Sonnet, or Perplexity's own experimental models, to process their query. It performs a live web search, synthesizes the findings, and presents a coherent answer with inline citations that link directly to the source webpages. Users can ask follow-up questions in a conversational thread, and the tool offers related suggested questions to dive deeper into a topic. It also provides the option to view the raw search results in a traditional list format alongside the AI-generated summary.
What sets Sonar apart is its model-agnostic, user-choice-centric design. Unlike standard AI search tools that operate as a black box with a fixed backend model, Sonar puts the decision in the user's hands, acknowledging that different models have different strengths in coding, creative writing, or logical reasoning. Technically, it acts as a sophisticated orchestrator, routing the user's query through the selected model's API while managing the web search, synthesis, and citation pipeline in a seamless interface. This creates a unique hybrid between a traditional search engine and an interactive AI research assistant.
Ideal for researchers, students, journalists, and professionals who need to quickly gather and synthesize information from the web with a high degree of accuracy and source transparency. Specific use cases include academic literature reviews, competitive market analysis, fact-checking for content creation, and technical troubleshooting where understanding the provenance of information is critical. It is particularly valuable in industries like media, consulting, and education where evidence-based answers are paramount.
As a freemium product, Sonar offers robust functionality in its free tier, which includes a generous number of queries with access to powerful models. For heavy users requiring higher usage limits and priority access to the latest models, Perplexity offers a professional subscription plan.