Perplexity AI’s Deep Research Tool Is Almost As Good As OpenAI and It’s Free
After OpenAI, Perplexity AI joins the “deep exploration” movement. And he does it in a rather interesting way. Following in the footsteps of DeepSeek’s “reasoning” model, Perplexity is the first major AI provider to also offer a Deep Research feature for free users. In comparison, OpenAI’s Deep Research feature is only available with a $200/month Pro subscription.
Deep Research is a new AI feature that takes little time, but performs dozens of related searches, crawls hundreds of resources, and uses a reasoning model to provide step-by-step logic for each query. Of course, you can get similar results using tools like Copilot’s “Think Deeper” feature, but what sets Deep Research apart is that it combines all the information it collects into a comprehensive white paper-style report.
Deep Research is free for all logged in users, although you are limited to just five requests per day. But if you pay for Perplexity Pro (which costs $20 per month or $200 per year), you get up to 500 requests per month. OpenAI’s Deep Research feature is currently limited to 100 requests per month.
It’s not an apples to apples comparison though. ChatGPT’s deep exploration feature takes a long time to complete. Up to 20 minutes. He asks clarifying questions, shows all the complex steps of the process and at the end gives a very long report. Perplexity’s Deep Research feature is kind of a lite version. You get an answer in 2-4 minutes, so your results will naturally depend more on web searches and data collation than on in-depth interpretation by the AI. OpenAI is using its upcoming o3 reasoning model for Deep Research, but Perplexity hasn’t mentioned any details about the model used.
In Humanity’s Last Exam , a widely used artificial intelligence test consisting of more than 3,000 questions on a range of topics, Perplexity’s Deep Research scored 21.1% in accuracy, much higher than DeepSeek R1 (8.6%) and Gemini (7.2%). OpenAI’s Deep Research still leads the pack with a completion rate of 26.6%, but the silver medal here is respectable given the tool’s much lower barrier to entry.
So how does this affect how you can use Perplexity? Until now, AI chatbots have been limited to a lot of prompts. You ask a question and prompt again and again to get detailed answers. But with Deep Research, you can ask one question and be done with it. The more specific and detailed your invitation is, the better the bot’s research and reporting will be, but now the AI can provide you with pages of information in response to far fewer requests. Once your report is created, you can download it in PDF format.
In their blog, Perplexity provided many examples of the difference Deep Research can make. Where the default model might provide an overview with short answers, Deep Research will instead return with multiple paragraphs, detailed reports, and more extensive formatting.