AI Deep Research

Deep research that ends in a cited report, not a chat scroll

Ask a question, or attach your own documents alongside it. Scholarly's research agent investigates across the open web and your sources in the background, then returns a report with inline citations you can check, share, and keep working from.

One free lifetime AI Creation Credit to try it, no credit card. Paid plans refresh weekly: 15 AI Creation Credits on Premium, 80 on Laureate.

200,000+ people already create with Scholarly
Direct answer

What is the best AI for research?

It depends on the job. For a quick fact, a chat assistant with web search is enough. For a question that needs many sources read and compared, a deep research agent fits better. Scholarly's Deep Research is that kind of tool: it plans the investigation, searches the open web, reads the files you attach, and returns a report with inline citations you can check.

  • How it automates the work: It asks a few clarifying questions when the goal is unclear, then searches, reads, and cross-checks sources and writes the report in the background. On paid plans, Extended research puts several researchers on the question in parallel, with a fact checker and a reviewer, for 2 AI Creation Credits.
  • What it reads: The open web, plus your own material: 1 attached source on Free, up to 10 on Premium, and up to 20 on Laureate. Evidence filters steer the search towards papers and journals, handbooks and manuals, news and industry sources, or government and policy publications.
  • What you get: A cited report, a brief, a comparison table, or slides, at one of three depths: a one-page brief, a decision memo, or a full investigation. The finished report stays open for follow-up questions.
  • What it is not: A paper database. It reaches open-access journals, preprints, and public reports, and it reads a paywalled paper only when you upload your own copy. It can also misread a source, so open the citations you plan to rely on.

Use Deep Research when you need a cited report on a question that takes many sources. Use a library database to search subscription journals, and a reference manager to keep a long-term source library.

Updated November 2026

How it works

From question to cited report in three steps

Deep Research runs as a background agent, so you can keep working while it reads.

01

Ask your question

Describe what you need to find out and what should come back - a cited report, a one-page brief, or a comparison table. Attach files, folders, or Google Drive documents if the answer should start from your own material.

02

The agent investigates

It clarifies the goal with a couple of focused questions if needed, then searches, reads, and cross-checks sources in the background - favouring the evidence classes you selected, such as peer-reviewed papers or government publications.

03

Read, verify, and follow up

The finished report cites its sources inline. Open the session to chat with the findings, question a claim, or queue a follow-up report in the same investigation.

What you get

Built for questions one search can't answer

Evidence reviews, option comparisons, and topic investigations that would otherwise cost you an afternoon of tabs.

Inline citations

Every report links the sources behind its claims, so you can verify the load-bearing ones instead of trusting a summary.

Your documents as ground truth

Attach up to 10 files, folders, or saved Scholarly items on Premium (20 on Laureate). The agent treats them as primary material and researches beyond them only to fill gaps.

Evidence filters

Steer the investigation towards papers and journals, handbooks and manuals, news and industry sources, or government and policy publications.

Depth to match the decision

Choose a one-page brief, a decision memo, or a full investigation - the same question, sized to how much rides on the answer.

Reports, briefs, and tables

Ask for the format the work needs: a cited report, an executive brief, a comparison table, or slides to present the findings.

A session, not a one-shot

Each investigation stays open. Chat with the finished report, challenge its reasoning, and add follow-up reports as the question evolves.

Extended research when it matters

Turn on Extended and a team takes the question instead of one researcher: parallel researchers, a scout checking what changed this year, an independent fact checker marking every claim confirmed, corrected, contested or unsupported, and a reviewer critiquing the draft. About 10–20 minutes, on paid plans.

Go deeper on a finished report

One press sends the team back out after newer data, primary sources, and the questions the first report could not settle. You get a second, complete report and keep the first.

The details

What is AI deep research?

Updated August 2026

Deep research is an AI workflow where an agent investigates a question over many sources instead of answering from a single model response. In Scholarly, you ask a question - optionally grounding it in your own documents - and the agent plans the investigation, searches and reads in the background, and returns a report with inline citations, sized as a one-page brief, a decision memo, or a full investigation.

  • Grounded two ways. It reads the open web and, when you attach them, your own PDFs, notes, folders, and Drive documents - your material comes first.
  • Cited by default. Claims link back to their sources, so the report is checkable rather than something you have to take on faith.
  • Runs in the background. Start it, keep working, and come back to a finished report you can chat with, share, or extend with follow-up runs.

The difference between deep research and a good chat answer is process. A chat reply is one pass over what the model already knows plus a quick search. A deep research run starts by pinning down what you actually need - the agent asks a couple of clarifying questions when the goal is ambiguous - then works through sources methodically, comparing what they say, noting where they disagree, and assembling the answer with the evidence attached.

That process matters most for questions where being roughly right isn't enough: what the trials actually showed, which vendor survives a real comparison, what a regulation requires, where the market has moved since last year. Those are questions you'd otherwise answer with an afternoon of open tabs and a document of pasted quotes.

Research grounded in your own documents

Most deep research tools only search the public web. Scholarly's runs inside your workspace, so an investigation can start from the material that actually defines your question: the internal report that needs external context, the vendor proposals you're comparing, the draft that needs its claims checked. Attach up to ten sources - uploaded files, workspace folders, saved pages, even earlier research sessions - and the agent treats them as primary evidence.

The evidence filters steer everything beyond your documents. Reviewing clinical or academic questions, favour papers and journals. Working on operational or technical material, favour handbooks and manuals. Tracking a market, favour news and industry sources. Checking compliance, favour government and policy publications.

When to use Deep Research instead of chat

Use chat when you need an answer now and one good pass is enough - a definition, a quick explanation, a summary of a document you've linked. Chat is immediate and conversational, and for most questions it's the right tool.

Use Deep Research when the question deserves evidence: evaluating options before a decision, reviewing what's actually known about a topic, checking a claim you're about to repeat, or building the cited document a colleague or client will read. The output is an artifact - a report you can share and return to - rather than a reply that scrolls away.

The report is the start, not the end

A finished investigation stays live in your workspace. Chat with it to question findings or pull out specifics, queue follow-up reports in the same session as the question sharpens, and hand the results onward - turn the report into slides for a briefing, a podcast for the commute, or study materials for a team that needs to learn it.

FAQ

Deep Research questions, answered

Is Scholarly's Deep Research free to use?

You can try it free: every account includes one lifetime AI Creation Credit, and a Deep Research run on the Auto model uses it. Paid plans refresh weekly - 15 AI Creation Credits on Premium and 80 on Laureate - shared across research, videos, podcasts, and other generated artefacts.

What sources does the research agent use?

The open web, plus anything you attach: uploaded files, workspace folders and pages, recordings, earlier research sessions, and Google Drive documents. Evidence filters let you favour peer-reviewed papers, handbooks and manuals, news and industry sources, or government and policy publications.

Are the reports actually cited?

Yes - reports cite the sources behind their claims inline, and the point of the format is that you can follow those links and verify anything load-bearing before you rely on it. Like any AI system, the agent can make mistakes, so check the citations that matter.

How long does a run take?

It depends on the depth you pick and how much reading the question demands. Runs happen in the background, so you don't wait on the page - start the investigation, keep working, and open the report when it's ready.

Can I control what the report looks like?

Yes. Choose the depth - one-page brief, decision memo, or full investigation - and say what to produce: a cited report, an executive brief, a comparison table, or slides. The agent asks a couple of clarifying questions up front when the goal needs sharpening.

How is this different from ChatGPT or Perplexity's deep research?

The investigation itself is similar; the difference is where it lives. Scholarly's runs inside a workspace that holds your documents, so runs can be grounded in your own material - and the finished report becomes a source you can chat with, cite, and turn into slides, podcasts, or training materials in the same place.

Which AI is best for research work?

Choose by the step you are on. To find and compare what many sources say about a question, use a deep research agent that returns a cited report, such as Scholarly's Deep Research. To understand one paper, use a tool that reads the PDF and answers from it. To search subscription journals or keep a reference library, use a library database and a reference manager, because Scholarly is neither.

Which tools automate multi-step online deep research?

Deep research agents do. They take one question, break it into searches, read the pages they find, compare the evidence, and write a cited report without you steering each step. Scholarly's Deep Research works this way, runs in the background, and also reads files you attach. Extended research, on paid plans, splits the question across several researchers and adds an independent fact check.

Can Deep Research search paywalled journals or a paper database?

No. It searches the open web, which reaches open-access journals, preprint servers, and government and institutional publications, and the Papers & journals filter steers it towards scholarly sources. It is not a subscription database and cannot open a paper behind a paywall. Upload your own copy of the PDF and the agent reads it as primary material.

Pricing

Ask the question you've been putting off researching

Your first investigation is free. Paid plans add weekly AI Creation Credits, bigger uploads, and stronger models.

Sample lecture

What one video lecture gets

Free makes a lecture of about 2 minutes in 720p with the Scholarly logo, to watch in Scholarly. Premium and Laureate make lectures up to 45 minutes in 1080p that you can download as an MP4, with the logo off if you choose.

Pricing shown in USD by default. Local currency may be available in app or at checkout.

Compare plans

Image creation · Beta

Shared across Image Studio and chat. Separate from AI Creation Credits. Free images are lifetime and do not reset. Paid allowances reset weekly without rollover. Limits may change during beta.

Free
2 lifetime
Premium
40/week
Laureate
80/week

AI chat

Premium and Laureate include unlimited AI chat on Auto. Chat starts on Best, and premium models such as GPT 6.2 Sol, Claude Sonnet 5.5, and Claude Opus 5.5 draw on a weekly premium usage allowance; when it runs out, chat keeps going on Auto until it resets. Chat never spends AI Creation Credits.

Free
100 total
Premium
Unlimited
Laureate
Unlimited

Premium chat models (Best)

Chat answers on Best, the strongest model, by default. Premium models draw on a weekly allowance that resets every Monday at 00:00 UTC; long threads, large files, background tasks, and Claude Opus 5.5 use it faster. When it runs out, chat keeps going on Auto until the reset.

Free
Not included
Premium
Weekly usage
Laureate
4× weekly usage

AI Creation Credits

AI Creation Credits cover videos, slides, documents, and other AI creations. Most models use 1 credit, a few use 2, and Frontier models use 2, 3, or 4. Image Studio and chat images have a separate weekly allowance.

Free
1 total
Premium
15/week
Laureate
80/week

Podcast length

Free episodes are about 2 minutes. Paid plans can set a target length up to 60 minutes.

Free
About 2 minutes
Premium
Up to 60 minutes
Laureate
Up to 60 minutes

Video lecture length

Free lectures with a length control are about 2 minutes. Paid plans can set a target up to 45 minutes. Short uses its own short video format.

Free
About 2 minutes
Premium
Up to 45 minutes
Laureate
Up to 45 minutes

Video lecture resolution

Free
720p
Premium
1080p
Laureate
1080p

Download lectures as MP4

Free
-
Premium
Included
Laureate
Included

Scholarly logo on lectures

Free
Always on
Premium
Your choice
Laureate
Your choice

File uploads

Free
1 total (8MB)
Premium
Unlimited (300MB)
Laureate
Unlimited (1 GB each)

PDF to flashcards

Free
32 pages
Premium
1000 pages
Laureate
3000 pages

Sources per creation

Free
Up to 1
Premium
Up to 10
Laureate
Up to 20

Practice questions

Free
5 total
Premium
Unlimited
Laureate
Unlimited

Practice exams

Free
1 total
Premium
Unlimited
Laureate
Unlimited

Voice mode

Free
5 min total
Premium
2 hr/day
Laureate
12 hr/day

Autocomplete

Free
500 words total
Premium
Unlimited
Laureate
Unlimited

Export to Anki

Free
-
Premium
Included
Laureate
Included

Support

Free
Standard
Premium
Priority
Laureate
Priority

What customers say

Scholarly has enhanced the quality of notes and questions I administer to my tutees. Insightful, to the point.

TD Georgas

TD Georgas

Tutor

I have very good experience with Scholarly it summarise me scientific text in hovel academic level with citation. I use this for any therapeutic area . It is great and useful tool.

Diana

Diana

Scholarly user

Scholarly has been a valuable tool for my studies. The AI-generated flashcards and intuitive features make organising and retaining information much easier.

Briana

Briana

Scholarly user

This app is great for studying for big test. Drop your PDF's in the system and it'll do the trick. You can organise it specifically for your needs.

Kelvin

Kelvin

Scholarly user

Your questions, answered

Is Scholarly free to use?

Yes! The free plan includes core study tools with lifetime allowances: 100 AI Chat messages, 500 autocomplete words, 5 quiz questions, 5 voice minutes, 1 practice exam, one free AI Creation Credit total, and 1 file upload. These allowances do not reset. Image Studio and chat also share 2 lifetime images during beta; this separate Free allowance does not reset and may change. Upgrade to Premium for 15 AI Creation Credits per week and unlimited uploads.

What uses my free AI Creation Credits?

AI Creation Credits are shared across flashcards, podcasts, videos, slides, AI Documents, Deep Research, and the Transcribe tool. The selected model determines the credit cost. Recording a meeting does not use an AI Creation Credit. Transcript and notes are included, up to 10 hours on paid plans.

Can I cancel any time?

Absolutely. There are no contracts or commitments. You can cancel your subscription at any time from your account settings, and you'll keep access until the end of your billing period.

What payment methods do you accept?

We accept all major credit and debit cards through Stripe. Pricing is displayed in USD by default, but local currency is available in the app.

Can I use Scholarly with a class or school?

Yes. For a school or organisation-wide rollout, contact us at hello@scholarly.so and we'll help you find the right setup.

What happens when I hit a free plan limit?

You'll see a prompt to upgrade. Your existing work is never lost. Limits only apply to new actions. Most Free allowances are lifetime limits. The separate beta image allowance includes 2 lifetime images on Free with no reset, 40 per week on Premium, and 80 per week on Laureate. Limits may change during beta. Premium includes 15 AI Creation Credits per week, podcasts up to 60 minutes, video lectures up to 45 minutes, unlimited uploads, chat, autocomplete, quizzes, and exams, plus more daily voice time.