Perplexity

Perplexity is an AI answer engine. You ask a question in plain language, it searches the live web, reads what it finds, and writes an answer with numbered citations next to each claim. Click a number and you land on the source. That citation habit is the whole product, and it is the reason people use it alongside a general chatbot rather than instead of one.

The workflow is simple enough that there is nothing to learn. Type a question, read the answer, follow the sources. Follow-up questions keep the thread's context, so you can start broad and narrow down without repeating yourself. Answers arrive in a few seconds for straightforward questions and take longer when the tool decides it needs to read more pages.

Beyond plain search there are three modes worth knowing. Pro Search runs more queries and reads more pages before answering, which is what you want for anything with a real answer buried in it. Deep Research goes further and produces a structured report over several minutes — useful for market questions, competitor scans, and topics where you would otherwise open twenty tabs. Spaces let you group related threads and upload files so the tool answers from your own documents alongside the web.

Perplexity does not run its own frontier model exclusively. Paid tiers let you pick which model handles a query, including current models from OpenAI, Anthropic and Google, so you are not locked to one vendor's reasoning style. For people who already pay for one chatbot, this is often the argument for adding Perplexity rather than switching.

Where it fits day to day: fact-checking a claim before you publish it, researching a topic you know nothing about, finding recent information a model's training data would not contain, and pulling together sources for a piece of writing. It is a research instrument, not a writing tool. Ask it to draft an article and you will get something serviceable and flat.

The free tier is not a crippled demo. You get unlimited standard searches with citations, which covers most casual use. Pro at $20/month adds far higher Pro Search limits, model selection, file uploads, image and video generation, and the Comet browser agent. Max at $200/month removes the ceilings and adds priority access — a price that only makes sense if your income depends on constant use.

The honest limitations. Answers occasionally cite a source that does not actually support the sentence attached to it, so for anything consequential you still have to click through. It is weak at long-form writing and creative work. The usage limits on Pro moved from a daily to a weekly ceiling, which means a heavy research day can hit the wall unexpectedly. And the pricing lineup has changed more than once in the past year, so check the current page rather than any article about it, including this one.

If you research for a living and you are tired of assembling your own sources, Perplexity earns its $20. If you mostly need something to write for you, it is the wrong tool and a general chatbot will serve you better.