Perplexity takes on ChatGPT and Gemini with new Deep Research AI that completes most tasks in under 3 minutes | Mint

Aravind Srinivas-led Perplexity AI has launched its personal Deep Research instrument, which it claims will assist customers save hours of time by performing in-depth analysis and evaluation on their behalf, whereas taking lower than 3 minutes for most tasks. Perplexity already supplied DeepSeek R1 on its platform, and the addition of Deep Research places it in direct competitors with AI giants like OpenAI and Google.

Perplexity says its Professional customers can have entry to 500 queries per day, whereas free customers will get a restricted variety of solutions. The new AI instrument is already accessible on Perplexity’s web site and the corporate says it can quickly be rolled out to iOS, Android and Mac variations of its apps.

To be able to use Deep Research, customers merely faucet on the new choice in the mannequin picker beneath the search field and click on on the new choice.

How does Perplexity’s Deep Research work?

Perplexity says that its Deep Research instrument performs dozens of searches, reads a whole lot of sources and causes what to do subsequent, much like how a human researcher may method a subject. As soon as the supply materials is full, the AI agent summarises all the data in a “clear and complete report” that will be exported both as a PDF or as a Perplexity web page to share with pals and household.

Perplexity says that Deep Research may also help present ‘expert-level evaluation’ on a variety of advanced subjects, together with finance, advertising and expertise. The corporate additionally claims that Deep Research can act as a ‘private marketing consultant’ in areas corresponding to well being, product analysis and journey planning.

ChatGPT vs Perplexity vs Google: How do the three Deep Research brokers fare?

By its personal admission, Perplexity lags behind ChatGPT Deep Research in the Humanity’s Final Examination accuracy rankings. Nonetheless, Perplexity claims that its newest AI providing beats the likes of Gemini Pondering, o3-Mini, Grok 2 and Claude 3.5 Sonnet on the identical benchmark.

Whereas Perplexity’s Deep Research might fall wanting ChatGPT in phrases of accuracy, Srinivas claims in a put up that the new AI instrument is “an order of magnitude quicker and cheaper” than the OpenAI providing. Notably, ChatGPT Deep Research solely provides 100 queries to its Professional customers, in comparison with Perplexity’s 500 queries/day for paid customers and restricted queries free of charge customers.