Perplexity kept the AI competition flowing on Monday with the debut of Deep Research, a service available via the Perplexity application that allows users to generate “in-depth research reports on any topic,” according to its social media account post.
Perplexity notes that the newly released Deep Research solution combines and leverages its query data via iterative search analysis, covering results from advanced web search, reasoning, and coding, therefore creating its end reports.
The Perplexity Deep Research service is free to all platform users. The firm notes that non-subscriber users can make up to 5 queries per day, and Pro users can make up to 500 queries per day.
Introducing Deep Research on Perplexity.
Deep Research lets you generate in-depth research reports on any topic.
Available to everyone for free—up to 5 queries per day for non-subscribers and 500 queries per day for Pro users. pic.twitter.com/obovx7YEUF
— Perplexity (@perplexity_ai) February 14, 2025
The Deep Research service is available via Perplexity’s mode selector in the search box function. Using this, a professional can iteratively search and read documents to help the user create an action plan for researching a new topic.
After collecting source material on a chosen subject matter, the Deep Research program can share AI-powered insights as a PDF or document report that the user can share among colleagues.
Shots Fired at AI Competitors
In its X thread, Perplexity explained that the Deep Research service “surpasses leading models in performance,” highlighting how the solution scored 93.9% accuracy on a SimpleQA benchmark. The company also keenly noted that Deep Research achieved 21.1% on Humanity’s Last Exam, “outperforming Gemini Thinking, o3-mini, o1, [and] DeepSeek-R1. ”
In an official statement, Perplexity added:
In addition to attaining high scores on industry benchmarks, Deep Research on Perplexity completes most tasks in under 3 minutes (and we’re working to make it even faster).
Perplexity is also seemingly keen to sell the Deep Research product to enterprise customers. The form continues by explaining the services’ analysis can help various end users, including finance, marketing, health, and tech.
Interestingly, the news comes following OpenAI’s debut of its own deep research service, an advanced agentic assistant. Deep Research by OpenAI operates independently, adapting its methods based on findings and making autonomous decisions regarding relevance and importance. This AI agent can analyze hundreds of online sources, adjust its research strategy according to what it discovers, and produce comprehensive reports in just minutes—tasks that would typically take human analysts many hours to complete.
OpenAI’s solution can pursue goals independently, assign research tasks, and return later to find a well-documented report complete with clear citations. Essentially, this represents a fundamental shift from viewing AI merely as a tool to recognizing it as an autonomous assistant.
The competition rises. If vendors and champions are to be believed, AI is poised to solve many business challenges, and with deep research tools becoming more widespread, the influence will soon be felt.
