Perplexity AI Labs is the most underused feature on the entire platform. Most people open Perplexity, ask a question, read the answer, and close the tab. They never discover that the same tool can take a single prompt and turn it into a complete spreadsheet, a working dashboard, a formatted research report, a presentation with slides, or a simple web app, all without any coding and without switching to another tool. This guide covers exactly what Perplexity AI Labs does, how to use it step by step, what it can and cannot build, and who gets the most practical value from it in 2026.
Table of Contents
What Is Perplexity AI Labs?
Perplexity AI Labs is a project-building mode within Perplexity that goes well beyond search and research. While standard search gives you a cited answer in seconds and Deep Research produces a multi-source report in two to five minutes, Labs is designed for longer, more complex tasks that produce finished, downloadable deliverables. Perplexity officially describes it as a mode that invests ten minutes or more of autonomous work using deep web browsing, code execution, chart creation, and file generation to complete your project.
In practice this means you can type a prompt like “Create a competitive analysis report comparing the top five project management tools in India” and Perplexity Labs will research the topic, structure the findings, generate charts, and produce a formatted, downloadable document. The difference from simply asking Perplexity a question is the output. Labs does not just answer. It builds.
Perplexity Labs was originally launched in May 2025 and has since been renamed “Create files and apps” in the platform’s interface, accessible via the “+” menu next to the search bar. The feature sits between Deep Research (fast, thorough reports) and Perplexity Computer (the Max-tier autonomous agent). Labs is available to all Pro, Education Pro, Max, Enterprise Pro, and Enterprise Max subscribers. It is not available on the free plan.
The outputs Labs can generate include documents (downloadable as PDF or DOCX), spreadsheets (downloadable as XLSX), presentations (downloadable as PPTX or HTML), web applications (deployed to a shareable URL), dashboards with interactive charts, and code files with supporting data assets. All files created during a Labs session are organised in an Assets tab from where they can be previewed, downloaded, or shared directly.
How Perplexity AI Labs Works?
The Autonomous Work Session
When you activate Labs and submit a prompt, Perplexity does not return a result in a few seconds. It begins an autonomous work session that typically runs for ten to thirty minutes depending on the complexity of what you asked for. During this session, the system combines several capabilities simultaneously: it browses the web for research and data, executes code to process and structure that data, generates charts and visual elements, and writes the files required to produce the finished output.
You can watch the work happen in real time. Labs shows you a Tasks pane on the left side of the screen that updates with each step it is completing, and a Sources pane that shows which pages it is pulling from as it researches. This transparency is useful for catching early on if Labs has misunderstood your prompt, so you can cancel and resubmit a clearer version rather than waiting fifteen minutes for an output that is not what you needed.
What Gets Built in the Assets Tab
Once Labs completes a session, every file it created appears in the Assets tab within that thread. For a report, this might include the main DOCX or PDF document, PNG images of charts embedded in the report, CSV files of the underlying data, and Python scripts used to generate the visualisations. For a web app, it includes the deployed HTML application and the code files behind it. You can download any individual asset, download everything as a zip file, or share the thread link with colleagues who can then access the same files.
The App tab appears separately when Labs has built an interactive application or dashboard. Perplexity deploys this to a dedicated URL that you can bookmark, share, or revisit without returning to the original thread. This makes it genuinely useful for building lightweight internal tools that colleagues can access without any setup.
How Labs Relates to Deep Research and Pro Search
As of 2026, Perplexity has integrated Labs-style file creation into both Deep Research and Pro Search modes as well. You can now prompt “Create a presentation on the AI agent market” directly within a Deep Research or Pro Search thread and Perplexity will generate the file as part of that session. This means file creation is no longer exclusively a Labs workflow. Labs remains the dedicated mode for longer, more complex project builds where the output requires substantial code execution, multiple file types, or interactive application deployment.
Think of it this way: Pro Search answers your question. Deep Research writes you a thorough report with citations. Labs builds you a finished, professional deliverable that you can hand to a client, present in a meeting, or deploy as a working tool.
How to Use Perplexity AI Labs: Step-by-Step Guide
Step 1: Access Labs from the Search Bar
Log in to your Perplexity Pro, Education Pro, or Max account at perplexity.ai. Click the “+” icon to the left of the main search bar. A small menu appears with mode options: Deep Research, Model Council, Create files and apps, and Learn step by step. Click “Create files and apps” to activate Labs mode. The search bar updates to reflect that Labs is now active.
Labs is available on the web interface, iOS app, and Android app. If you are on mobile, the “+” menu is accessible in the same position relative to the search bar. File preview and download work on both platforms, though complex interactive apps are easier to review on desktop.
Step 2: Write a Clear, Specific Project Prompt
The single most important factor in getting a useful Labs output is prompt quality. Labs takes your prompt literally and invests ten or more minutes working on it. A vague prompt produces a vague deliverable. A specific, well-scoped prompt produces something genuinely usable.
A weak prompt: “Make a report about marketing.” A strong prompt: “Create a research report on digital marketing trends for small businesses in India in 2026. Include sections on social media marketing, SEO, WhatsApp marketing, and content marketing. Add data visualisations where relevant. Format it as a professional report I can share with a client.”
Effective Labs prompts specify the output format you want (report, spreadsheet, dashboard, presentation, web app), the topic with clear scope, the audience or purpose, and any specific sections or data points you want included. The more detail you give upfront, the less likely you are to need to resubmit.
Step 3: Monitor the Work Session
After submitting, Labs begins its autonomous session. The Tasks pane on the left of the screen updates in real time, showing you each step as it completes: web searches it is running, code it is executing, charts it is generating, and files it is creating. Do not close the tab during this process.
If you see in the first two to three minutes that Labs is clearly working on something different from what you intended, click “Stop” to cancel the session before it wastes more time. Rewrite your prompt with more specific instructions and resubmit. If the direction looks correct, let it run to completion. Most sessions finish within ten to twenty minutes for standard report or spreadsheet tasks. Interactive apps and dashboards can take longer, up to thirty minutes for complex builds.
Step 4: Review and Download Your Assets
When the session completes, open the Assets tab that appears at the top of the Labs thread. Click any file to preview it inline before downloading. For documents and presentations, the preview shows the formatted output as it will appear when downloaded. For code files and CSVs, you see the raw content.
Download the files you need individually or as a zip archive. If the output is mostly correct but needs adjustments, use a follow-up prompt in the same thread to request specific changes rather than starting a new Labs session. For example: “The report looks good. Please add a conclusion section summarising the three most important trends, and update the chart colours to blue and grey.”
Perplexity supports version history within each thread, so you can review previous versions if a follow-up prompt changes something you preferred in an earlier iteration.
Step 5: Share or Deploy Your Output
For standard files like reports, spreadsheets, and presentations, download them and use them in your existing workflow. DOCX files open in Microsoft Word or Google Docs for further editing. XLSX files open in Excel or Google Sheets. PPTX presentations open in PowerPoint or Google Slides.
For web apps and dashboards, Labs deploys the output to a dedicated shareable URL. Click “Open App” in the App tab to access the deployed version. Copy the URL to share with colleagues, embed in a project brief, or bookmark for ongoing reference. The app remains accessible at that URL without requiring anyone to have a Perplexity account.
Step 6: Iterate with Follow-Up Prompts
Labs supports follow-up editing within the same thread. After reviewing your output, you can ask for specific changes, additional sections, different data visualisations, or reformatting without starting from scratch. Keep follow-up prompts as precise as possible. Asking for broad changes like “make it better” tends to produce unpredictable results. Asking for specific changes like “replace the bar chart in section three with a line chart showing the trend over time” works reliably.
One important practical note: Labs, like most AI coding tools, can sometimes struggle with follow-up changes that require significant reworking of generated code. If an app breaks after a follow-up prompt, ask Labs to fix only the specific problem with minimal changes rather than asking it to rewrite the whole thing. Minimal targeted fixes tend to produce more stable results than broad rewrites.
Key Benefits of Perplexity AI Labs
Turns Research Directly Into a Deliverable
The most practical benefit of Perplexity AI Labs is that it collapses two separate phases of work into one. The traditional workflow for producing a research report or competitive analysis involves researching the topic across multiple sources, taking notes, structuring the findings, formatting the document, and adding charts. Each of those steps normally happens in a different tool. Labs does all of it in a single session from a single prompt. The output is not a rough draft that requires substantial cleanup. For most standard use cases, it is a structured, formatted deliverable that needs only light editing before it is ready to use.
No Coding Required for Functional Apps and Dashboards
Building an interactive dashboard or a simple web application used to require either a developer or significant time learning a no-code tool. Labs lets non-technical users describe what they want in plain language and receive a working, deployed app in under thirty minutes. A marketing manager can prompt “Build a dashboard showing our campaign performance metrics with filters for date range and channel” and receive an interactive HTML application at a shareable URL. The code behind it is available in the Assets tab for anyone who wants to inspect or extend it. This is not a replacement for production-grade development, but for internal tools, presentation demos, and lightweight applications, it removes the technical barrier entirely.
All Assets Organised in One Place
Every file Labs generates during a session is saved in the Assets tab of that thread. Charts, CSVs, code files, documents, and images are all organised together and accessible from a single location. You can download them individually or as a zip. The thread itself remains in your Perplexity history and can be reopened at any time to access the files again or continue iterating. This organisation is more practical than it might sound: generating assets across multiple tools means hunting through downloads folders, email attachments, and cloud storage. Having everything from one project session in one place saves time when you need to revisit, share, or update the work.
Integrated with Research for Credible Outputs
Because Labs runs inside Perplexity’s research infrastructure, the content it produces is grounded in real web sources with citations attached where relevant. A report produced by Labs does not just contain generated text. It contains synthesised findings from actual sources, with citations included in documents by default. This makes the outputs more credible and more useful than equivalent outputs from a pure language model that generates content from training data alone. For reports and documents that will be shared externally or used in professional contexts, the citation foundation is a meaningful differentiator from similar tools that do not integrate live research.
Perplexity AI Labs vs Alternatives: Comparison Table
| Tool | What It Creates | Research Integration | Coding Required | Deployment | Monthly Cost |
|---|---|---|---|---|---|
| Perplexity AI Labs | Reports, spreadsheets, presentations, dashboards, web apps | Yes, live web with citations | No | URL deployment for apps | Included in Pro ($20/mo) |
| ChatGPT (Canvas + Code) | Documents, code, data analysis | Optional web search | No for docs, Yes for apps | No deployment | Plus $20/mo |
| Notion AI | Documents, databases, summaries | No live search | No | Within Notion only | $10/mo add-on |
| Gamma | Presentations, documents | Limited | No | Shareable link | Free / $15/mo |
| Google Gemini + Workspace | Docs, Sheets, Slides | Google Search integration | No | Within Google Workspace | Free / $19.99/mo |
| v0 by Vercel | Web apps and UI components | No | No for UI, Yes for custom | Vercel deployment | Free / $20/mo |
Who Should Use Perplexity AI Labs?
Freelancers and consultants who regularly produce research reports, competitive analyses, or client briefings will find Labs genuinely useful for the first-draft stage of deliverable production. A market research report that would normally take a full day of research, writing, and formatting can be produced in a thirty-minute Labs session and polished in another thirty minutes of editing. For consultants billing by the hour, this compression in production time has a direct financial value. The ability to produce professional-looking deliverables quickly also makes it practical for client proposals and exploratory research that might not justify a full day of manual work.
Students and academics working on presentations, data visualisation assignments, and structured research projects can use Labs to generate well-organised starting points. The research integration means the content is sourced rather than invented, and the citation inclusion in documents means the academic credibility of the output is higher than equivalent AI-generated text from tools without live search. For group projects where one person handles the research compilation, Labs significantly reduces that workload.
Small business owners and marketers who need internal tools, performance dashboards, or structured market research without a technical team behind them will get practical value from Labs’ app and dashboard generation. A business owner who needs a simple inventory tracking tool, a marketing team that needs a campaign performance dashboard, or a manager who needs a formatted competitor comparison report can all get usable outputs from Labs without involving a developer or spending hours in spreadsheet tools.
Content creators and bloggers building content around data-driven topics can use Labs to generate supporting charts, tables, and structured summaries that would otherwise require manual data processing. Producing a “State of [Industry]” report with formatted data visualisations, which typically requires pulling data from multiple sources and manually creating charts, becomes a single Labs prompt. This is particularly relevant for Edurancehub’s own content production workflow when covering AI tool comparisons, pricing analyses, and feature breakdowns.
External Resources
- Perplexity AI Official Site — Primary reference throughout
- Perplexity AI Pro Page — Pricing section
- Perplexity AI Deep Research Blog — Deep Research step section
- Perplexity AI Help Center: How It Works? — How it works section
- Interview Kickstart: Perplexity for Content Research — Research workflow section
- SimilarLabs: Perplexity AI Review 2026 — Comparison table reference
- Techyden: Perplexity Computer for Bloggers — Labs and automation section
- Incremys: Perplexity AI Statistics 2026 — 45% creator stat citation
- Google Scholar — Reference alongside Academic Focus mode
- Reddit — Reference in Reddit Focus mode section
Frequently Asked Questions
Is Perplexity AI Labs free to use?
Perplexity AI Labs is not available on the free plan. It requires a Pro subscription at $20 per month, an Education Pro subscription at $10 per month (or free for 12 months at qualifying institutions), or a Max subscription at $200 per month. Enterprise Pro and Enterprise Max subscribers also have access. If you are on the free plan and want to try Labs, upgrading to Pro is the entry point. For students who qualify for Education Pro, that plan provides Labs access at a significantly reduced cost. The Pro plan at $20 per month includes Labs access with a limited monthly query quota, while Max provides unlimited Labs usage with no cap on the number of projects or sessions.
What file types can Perplexity AI Labs create?
Labs can generate documents in PDF and DOCX format, spreadsheets in XLSX format, presentations in PPTX and HTML format, and web applications in HTML that are deployed to a shareable URL. Supporting assets generated during any Labs session, such as PNG charts, CSV data files, and Python scripts, are also available for download in the Assets tab. As of 2026, file creation is also supported directly within Deep Research and Pro Search modes, so you do not always need to activate dedicated Labs mode to generate a document or presentation. The file type support continues to expand with platform updates.
How long does a Perplexity AI Labs session take?
Most Labs sessions complete in ten to thirty minutes. Simple reports and spreadsheets typically finish in ten to fifteen minutes. More complex outputs involving interactive dashboards, web applications, or multi-section reports with substantial data processing can take twenty to thirty minutes. The Tasks pane shows you each step in real time, so you can monitor progress and estimate remaining time. Labs is designed for situations where you can step away from the task and return to a finished output rather than watching it generate line by line. If you need something in under five minutes, standard Pro Search or Deep Research will serve you better for that use case.
Can I edit the files that Perplexity AI Labs creates?
Yes, in two ways. Within the Perplexity thread, you can use follow-up prompts to request specific edits, additions, or format changes. Labs will update the output based on your instructions and track the version history so you can revert to an earlier version if needed. You can also download the files and edit them directly in your preferred application. DOCX files open in Microsoft Word or Google Docs, XLSX files open in Excel or Google Sheets, and PPTX files open in PowerPoint or Google Slides, all of which allow full manual editing. For web apps, the HTML code is available in the Assets tab and can be edited by anyone comfortable working with HTML and CSS.
How is Perplexity AI Labs different from Deep Research?
Deep Research produces a structured, cited research report by running dozens of iterative searches across hundreds of sources in two to five minutes. The output is a text-based report within the Perplexity interface that you can copy or export. Labs produces finished, downloadable files including documents, spreadsheets, presentations, dashboards, and web apps by running a longer autonomous session of ten to thirty minutes that combines research, code execution, and file generation. The practical distinction is that Deep Research is the right tool when you need a thorough, sourced answer or research brief quickly. Labs is the right tool when you need a finished deliverable in a specific file format that you can hand to someone else, publish, or deploy directly. For a full breakdown of how Deep Research works, see our Perplexity AI Deep Research guide.
Does Perplexity AI Labs work on mobile?
Yes. Labs is accessible on both the iOS and Android Perplexity apps via the “+” menu next to the search bar. You can submit prompts, monitor the work session, and download completed files from your phone. However, interactive app previews and complex dashboard outputs are more comfortable to review on a desktop browser where the screen real estate allows you to see the full output. For submitting prompts and reviewing completed reports and documents on the go, the mobile experience works well. File downloads from the mobile app save directly to your device’s storage, from where you can open them in the relevant app such as Microsoft Word, Excel, or a PDF viewer.
Final Thoughts
Perplexity AI Labs is genuinely one of the more capable features in the current AI tools landscape, and it is included in the Pro plan at $20 per month rather than locked behind a separate premium tier. The ability to go from a natural language prompt to a formatted report, a working spreadsheet, or a deployed interactive dashboard in under thirty minutes without writing code or switching tools is a real workflow improvement for the types of tasks it handles well.
It is not perfect. Complex follow-up edits to generated code can sometimes break the output rather than improving it, and the ten-to-thirty-minute session time makes it unsuitable for quick tasks where standard search or Deep Research is the better choice. But for producing finished first drafts of professional deliverables, it covers a lot of ground for the price.
The best way to understand what Labs can do for your specific workflow is to try it on one real project you have been putting off because it felt too time-consuming. Pick something with a clear output format, write a specific prompt, and let the session run. For more on the features surrounding Labs in the broader Perplexity ecosystem, our complete guide to using Perplexity AI covers the full platform, and our Perplexity AI plans breakdown explains exactly what is included at each subscription tier.
