Perplexity Deep Research Review: Advanced Deep Research, Plans and API [WTF Radar]

Scorecard for Perplexity Deep Research on a dark navy background

Perplexity Deep Research writes cited reports in minutes, with an Advanced Deep Research update that adds clarifying questions, a code sandbox and editable reports. Plans, Sonar API and alternatives.

Perplexity built its brand on answers with citations, so deep research was a short step. Its version is quicker than ChatGPT's or Gemini's and lives where many people already search.

Key Takeaways

  • Fast and citation-first. — Launched February 2025 with free daily queries and reports in two to four minutes.
  • Advanced Deep Research. — Clarifying questions, follow-ups mid-run, a code sandbox, document uploads and reports that stream into an editable file.
  • Model by plan. — Max subscribers get Claude Opus 4.6 Thinking; Pro gets a lighter top model, rolling out gradually.

What does Perplexity Deep Research do?

Perplexity launched Deep Research on February 14, 2025, with five free queries a day and reports in about two to four minutes (Engadget). The Advanced Deep Research update adds clarifying questions, follow-ups during a run, a progress view, key findings before the final report, a code sandbox, document uploads and broader browsing; Max subscribers get Claude Opus 4.6 Thinking (Perplexity Help Center).

Perplexity says it matches or beats competitors on benchmarks but names none. In the Columbia Tow Center's March 2025 test of eight AI search tools, Perplexity had the lowest citation failure rate at 37% (Nieman Lab); that test covered search, not deep research.

How does Perplexity Deep Research score?

Editorial scores from published documentation and announcements through October 31, 2026, not benchmark results. Total: 15/20, on the same criteria as the Best AI Research Tools guide.

CriterionScoreWhy
Depth4/5Advanced mode reads more sources and cross-references them; still faster and lighter than ChatGPT or Gemini.
Citations4/5Built around inline citations; best among eight tools in the Tow Center's 2025 citation test, still wrong 37% of the time.
Reach3/5Web and uploaded documents; fewer work-app connectors documented.
Access4/5Free tier, Pro $20 a month, Sonar Deep Research API.

How is Perplexity Deep Research priced?

Included in Perplexity plans. As published in April 2026: Free; Pro $20 a month or $200 a year; Max $200 a month; Enterprise Pro $40 and Enterprise Max $325 per seat a month. The Sonar Deep Research API is billed at $2 per million input tokens and $8 per million output tokens plus citation, reasoning and search fees (Finout).

Who is Perplexity Deep Research for?

  • Fits: people who already search with Perplexity and want quick, cited reports; developers who want a research API priced by tokens.
  • Look elsewhere: for the deepest runs (ChatGPT, Gemini) or research over company data (Claude Research, Copilot Researcher).

What are the Perplexity Deep Research alternatives?

Sources

Is Perplexity Deep Research free?

Perplexity launched Deep Research in February 2025 with free daily queries, and the free plan still includes limited research. Pro and Max raise the limits and use stronger models. Perplexity changed usage limits during 2026, so check current caps in the product.

What is Advanced Deep Research?

It is Perplexity's update to its Research feature: clarifying questions before a broad query, follow-up questions while research runs, a progress view, a code sandbox for calculations, document uploads, and reports that stream into an editable, shareable file.

Is Perplexity Comet the same as Deep Research?

No. Comet is Perplexity's agentic web browser. Deep Research is the research mode inside Perplexity's answer engine. We review Comet separately.

Explore More

Ready to Find the Right AI Tools?

Browse our data-driven rankings to find the best AI tools for your team.