NativeAIHub
All Tools/
ChatGPT Deep Research logo

ChatGPT Deep Research

other

by openai

An agentic research capability that autonomously browses the web for 5 to 30 minutes, synthesizing hundreds of sources into comprehensive cited reports.

Key features

Autonomous Web Browsing
Structured Reports with Inline Citations
Real Time Progress Tracking
File and Spreadsheet Upload
Full Model (o3) and Lightweight (o4-mini)
MCP/App Connections (February 2026)
Pricing

Free tier available, Plus at $20/mo

Best For

Researchers needing comprehensive literature reviews with hundreds of cited sources

Verdict

Produces comprehensive reports that would take hours to compile manually, complete with structured sections and inline citations from hundreds of sources

What it does

Autonomous Web Browsing

The agent autonomously browses the web for 5 to 30 minutes, planning its own research strategy, searching multiple angles, reading full articles, and following leads without human intervention.

Learn more

Structured Reports with Inline Citations

Every report is organized with clear sections, summaries, and inline citations linking back to the original sources. You can verify any claim by clicking through to the source page.

Learn more

Real Time Progress Tracking

A sidebar shows what the agent is doing in real time: which sources it is reading, what searches it is running, and what reasoning steps it is taking. You can monitor the entire investigation as it happens.

Learn more

File and Spreadsheet Upload

Upload PDFs, spreadsheets, documents, and other files to give the agent additional context. It incorporates your data into its research, building on what you already know.

Learn more

Full Model (o3) and Lightweight (o4-mini)

Two research modes. The full model uses o3 for the most thorough investigations (10 to 30 minutes). The lightweight model uses o4-mini for faster, simpler research tasks (5 to 15 minutes).

Learn more

MCP/App Connections (February 2026)

Connect Deep Research to external apps and data sources via MCP. The agent can access private data, internal tools, and custom services during its research.

Trusted Site Restriction

Restrict the agent's web searches to specific trusted domains. Useful for ensuring research draws only from verified, authoritative sources.

Image and PDF Interpretation

The agent can read and interpret images, charts, graphs, and PDF documents it encounters during web browsing, not just text content.

Interruptible with Follow Up Prompts

Pause the research at any time, provide additional context or redirect the investigation, and the agent adjusts its strategy accordingly.

Pricing

Free

Free

5 lightweight Deep Research queries per month using o4-mini. Shorter research sessions (5 to 15 minutes). Good for occasional research needs.

  • 5 lightweight queries/month
  • o4-mini model
  • 5 to 15 minute research sessions
  • Structured reports with citations
  • Real time progress tracking
Best Value

Plus

$20/month

10 full model queries plus 15 lightweight queries per month (25 total). Full model uses o3 for the most thorough research. The best value for regular research needs.

  • 10 full model (o3) queries/month
  • 15 lightweight (o4-mini) queries/month
  • 25 total queries/month
  • 10 to 30 minute research sessions (full)
  • File upload for context
  • All report features

Pro

$200/month

125 full model queries plus 125 lightweight queries per month (250 total). The highest research capacity. MCP integration, trusted site restriction, and priority access.

  • 125 full model (o3) queries/month
  • 125 lightweight (o4-mini) queries/month
  • 250 total queries/month
  • MCP/app connections
  • Trusted site restriction
  • Priority access to new features

Pros & Cons

Pros

  • Produces comprehensive reports that would take hours to compile manually, complete with structured sections and inline citations from hundreds of sources
  • Inline citations from hundreds of sources let you verify every claim and dig deeper into any specific finding
  • Real time progress tracking shows exactly what the agent is doing, so you can monitor the investigation and redirect if needed
  • Available on the Free tier (5 lightweight queries per month), making it accessible for occasional research needs
  • MCP integration (February 2026) allows the agent to access private data and internal tools during research
  • Can process uploaded files (PDFs, spreadsheets, documents), so the agent builds on your existing knowledge rather than starting from zero

Cons

  • Slow by design: each research session takes 5 to 30 minutes, making it unsuitable for time sensitive questions that need immediate answers
  • Very limited queries on Free (5/month) and Plus (25/month) tiers; heavy researchers will hit limits quickly
  • Pro plan at $200/mo is expensive if Deep Research is your primary use case
  • The lightweight model (o4-mini) is noticeably less thorough than the full model (o3), producing shorter reports with fewer sources
  • Cannot access paywalled content; the agent can only read freely accessible web pages
  • No API access for Deep Research specifically; developers cannot integrate this capability into their own applications

How to get started

1

Open ChatGPT

Go to chatgpt.com or open the ChatGPT app. Deep Research is available on all plans, including Free (5 lightweight queries per month).

chatgpt.com
2

Select Deep Research

Click the Deep Research button in the model selector or tools menu. On Plus and Pro plans, you can choose between the full model (o3, more thorough) and lightweight model (o4-mini, faster).

3

Describe your research question

Be extremely specific. Instead of "Tell me about electric vehicles," ask "Analyze the competitive landscape of electric vehicle manufacturers in Europe, focusing on pricing, battery technology, charging infrastructure partnerships, and market share trends from 2024 to 2026." Specificity produces dramatically better results.

4

Upload context files (optional)

Attach PDFs, spreadsheets, or documents that contain relevant data or context. The agent will incorporate your files into its research, building on what you already know rather than duplicating it.

5

Monitor progress and redirect if needed

Watch the real time progress sidebar as the agent browses, reads, and reasons. If it goes down an unproductive path, send a follow up message to redirect the investigation. You can also add new sources or constraints mid research.

6

Review the report

When the agent finishes (5 to 30 minutes), review the structured report. Check the inline citations for key claims. Ask follow up questions to drill deeper into specific findings.

Deep dive

Detailed guides with comparisons, tips, and visuals for each feature.

How Deep Research Works

The agentic research loop: browse, read, reason, pivot, synthesize. Full model (o3) vs lightweight (o4-mini). The 5 to 30 minute execution cycle explained.

All plans (limits vary)·2 min read

Use Cases and Best Practices

Market research, competitive analysis, academic literature review, due diligence, and technical deep dives. How to get the best results from Deep Research.

All plans·2 min read

Deep Research vs Perplexity Pro vs Google Deep Research

How ChatGPT Deep Research compares to Perplexity's deep research capabilities and Google's equivalent feature across depth, speed, citations, and pricing.

All plans·1 min read

Get notified about updates

We'll email you when this tool's pricing or features change.

Last updated: 2026-02-21