Every ranking page about AI research tools answers the wrong question. They list six or seven apps, rank them by vague “accuracy scores,” and quote prices in dollars — as if their readers never hit a paywall mid-semester and never had to explain a card transaction in rupees to their parents. The question Indian students and researchers actually ask is much more practical: which of these tools fits the task I have right now, what will it actually cost me in rupees, what do I lose on the free tier, and will my Indian card even work on their checkout?
This guide answers exactly that, for the three tools that matter most in 2026: SciSpace, Elicit, and Consensus. Every price below was checked on the vendors’ own pricing pages on October 23, 2026, every USD figure is converted at ≈₹94.4/USD, and where a primary source was silent, the article says so instead of guessing.
One clarification before we start: this is about discovering and understanding published research — finding papers, extracting data, answering evidence questions. If your need is organizing your own notes and PDFs, our Gemini Notebook guide covers that different job.
The short answer
If you don’t want to read the whole comparison, here is the honest verdict:
- Best all-rounder for Indian students: SciSpace — the free Basic plan (100 AI credits/month) genuinely covers reading and understanding papers; Premium is $12/month billed annually (≈₹1,130/month) with unlimited chats, summaries, and exports.
- Best for systematic literature reviews: Elicit — free for unlimited paper search across 138M+ papers; the Plus plan at $11/month billed annually (≈₹1,040/month) unlocks structured extraction; Pro at $39/month (≈₹3,680/month) adds the systematic-review workflow that screens 5,000 papers.
- Best for quick evidence answers: Consensus — free tier gives 10 AI-powered Pro messages and 3 Deep Reviews a month; Pro is $12/month billed annually (≈₹1,130/month) with unlimited Pro messages.
- The ₹0 stack that covers most needs: Elicit’s free unlimited search + SciSpace’s free 100 credits/month + Consensus’s free 10 Pro messages/month. Between the three free tiers, a student can run a serious literature search without paying anything.
- The payment reality: all three bill in USD only. None mentions INR billing or UPI anywhere. You need an international-enabled card, and expect roughly 2–3.5% forex markup.
The three tools, and the three different jobs they do
The single most useful thing to understand is that these are not three versions of the same product. They solve three different problems:
| SciSpace | Elicit | Consensus | |
|---|---|---|---|
| Core job | Read, understand, and write about papers | Extract structured data across many papers | Answer questions with cited evidence |
| Think of it as | Your research reading desk | Your literature-review spreadsheet | Your evidence Q&A engine |
| Best for | Understanding one paper deeply; writing | Systematic reviews; comparing studies | “Does X actually work?” questions |
| Free tier strength | 100 AI credits/month, limited everything | Unlimited search + summaries + chat | 10 Pro messages + 3 Deep Reviews/month |
| Starts at (annual) | $12/mo (≈₹1,130) | $11/mo (≈₹1,040) | $12/mo (≈₹1,130) |
Pick the tool by the task, not by the ranking. A PhD student doing a systematic review needs Elicit; a master’s student trying to understand five dense papers needs SciSpace; a researcher checking whether the evidence supports a claim needs Consensus. Most serious researchers end up using at least two.
SciSpace — the reading and writing desk
SciSpace is the broadest of the three: literature search, chat with PDFs, a paraphraser, a citation generator, an AI writer, and an AI agent that can run research tasks. If you have ever stared at a 20-page paper wondering what the methodology section actually did, SciSpace’s “explain this paper” workflow is the reason to open it.
What it costs (verified on scispace.com/pricing, October 23, 2026)
| Plan | Price | ≈ in rupees | What you get |
|---|---|---|---|
| Basic | $0 | ₹0 | 100 AI-agent credits/month, limited literature-review search, limited chats with papers and PDFs, limited summaries and citations, no exports |
| Premium | $12/mo, billed annually | ≈₹1,130/mo (≈₹13,600/year) | 1,200 credits/month, unlimited literature-review search, unlimited chats/summaries/paraphrasing, unlimited exports (RIS, CSV, BIB, Excel, XML) |
| Advanced | $70/mo, billed annually | ≈₹6,610/mo | 5,500 credits/month, Deep Review model access, everything in Premium |
| Teams | from $8/user/mo | ≈₹755/user/mo | Pooled credits, admin controls |
The free-tier truth
The Basic plan’s 100 credits a month is enough to chat with a handful of papers and run a few summaries — genuinely useful for a student working through a reading list. The trap is the no exports rule: you can find and read papers on the free plan, but you cannot export your literature review to a spreadsheet or citation manager. The moment your workflow needs RIS/BIB exports for Zotero or Mendeley, you are paying.
The Indian angle: SciSpace started in Bengaluru
This is the detail no global ranking page mentions. SciSpace began life as TypeSet, founded in Bengaluru in 2016 by Saikiran Chandha and Shanu Kumar as a platform that formatted academic papers for journal submission — the tedious template work every researcher hates. It raised $850K in a pre-Series A round in 2017, and over the years rebuilt itself around AI: first the paper-explainer that made it famous with students, now a full agent-based research platform. The rebrand from TypeSet to SciSpace reflects that shift, but the founding story is Indian — worth knowing when a global listicle treats it as just another Silicon Valley app.
One caution: a secondary blog claims SciSpace offers a 30% student discount for India with a code (“IND30”, bringing Premium to about $8.40/month). We could not verify this anywhere on SciSpace’s official pricing or help pages. Treat it as unconfirmed — check the pricing page yourself before counting on it.
Elicit — the extraction machine
Elicit does one thing better than anyone: it reads hundreds of papers and pulls structured answers into a table. Ask “what sample sizes did these 50 studies use?” and Elicit fills a column. For systematic literature reviews — the kind with PRISMA flowcharts that PhD students dread — this is the tool.
What it costs (verified on elicit.com/pricing, October 23, 2026)
| Plan | Price | ≈ in rupees | What you get |
|---|---|---|---|
| Basic | $0 | ₹0 | Limited Research Agent + Research Reports usage; unlimited search across 138M+ papers, unlimited summaries, unlimited chat with papers, Zotero import |
| Plus | $11/mo, billed annually ($132/year) | ≈₹1,040/mo (≈₹12,460/year) | Standard agent/reports usage, 5 table columns at a time, 500K clinical-trials search, API access |
| Pro | $39/mo, billed annually ($468/year); $49/mo month-to-month | ≈₹3,680/mo (≈₹44,180/year) | 5× usage, systematic-review workflow screening 5,000 papers, 20 columns, reports from 135 data sources |
| Scale | $89/mo, billed annually | ≈₹8,400/mo | 9× usage, 30 columns, figure extraction, team collaboration |
The free-tier truth
Elicit’s free tier is the most generous of the three for the core job: unlimited search, unlimited summaries, unlimited chat with papers. The limits bite only when you need the agent to do heavy extraction work — the “Research Agent” and “Research Reports” quotas are where Basic runs out. For a student doing a course-level literature survey, the free tier may be all you ever need. The jump to Pro ($39/month) is priced for funded researchers and labs, not students — at ≈₹44,000 a year, it costs more than many Indian PhD monthly stipends, so be sure you need systematic-review scale before paying.
Worth knowing
Elicit’s corpus is 138M+ papers, built on Semantic Scholar’s database — so its coverage of published literature is genuinely broad. Its enterprise tier advertises “no training on your data by default” and PRISMA-grade screening accuracy, which tells you who the real customer is: research teams and institutions, with students as the funnel.
Consensus — the evidence answer engine
Consensus answers a different question from the other two: not “find me papers” but “what does the evidence say?” Ask whether a treatment works or an intervention is effective, and it returns a synthesized answer with citations to real studies — plus a “consensus meter” showing where the evidence leans. For quickly checking a claim before you build an argument on it, nothing else does this as directly.
What it costs (verified on consensus.app/pricing, October 23, 2026)
| Plan | Price | ≈ in rupees | What you get |
|---|---|---|---|
| Free | $0 | ₹0 | Basic paper search (no AI analysis), 10 Pro messages/month, 3 Deep Reviews/month |
| Pro | $12/mo billed annually ($144/year); $20/mo month-to-month | ≈₹1,130/mo (≈₹13,600/year) | Unlimited Pro messages, 15 Deep Reviews/month, unlimited access to all research tools |
| Deep | $45/mo billed annually ($540/year); $65/mo month-to-month | ≈₹4,250/mo | Everything in Pro, 200 Deep Reviews/month |
The free-tier truth
Ten Pro messages and three Deep Reviews a month is enough to test whether Consensus fits your workflow — and to answer a few real questions a month. It is not enough for sustained research; a student running daily evidence checks will burn through it in the first week. Note that “Study Snapshots,” which older reviews mention as a free-tier feature, no longer appear anywhere on the current pricing page — don’t rely on outdated reviews for the free limits.
Consensus also offers up to 40% off for students and faculty with a valid school email (there is a claim banner on the pricing page). Whether Indian university email addresses qualify was not verifiable without signing up — worth trying with your institutional email before paying full price.
The money truth: why everything is in dollars
We checked all three pricing pages, FAQs, and billing sections. Not one mentions INR billing, UPI, netbanking, or any India-specific payment method. Every checkout is USD through a card processor.
What that means in practice for an Indian student or researcher:
- You need an international-enabled debit or credit card. Domestic-only cards will fail at checkout.
- Your bank converts at its own rate and typically adds a 2–3.5% forex markup — so that $12/month plan costs closer to ₹1,160–₹1,170 in practice, not the ₹1,130 mid-market figure.
- Annual billing saves real money on all three (SciSpace Premium $12 vs $20 monthly; Elicit Plus $11 vs higher monthly rates; Consensus Pro $12 vs $20 monthly) — but only commit annually after a month of actual use.
- Our guide to how Indians actually pay for AI tools covers this USD-billing wall across the whole AI tool landscape — research tools are no exception.
The academic-integrity section nobody writes
Here is the honest part. These tools are superb for discovery: finding papers you would never have found, understanding methods sections, extracting data for your review tables. That is legitimate research assistance, and no university objects to a better search engine.
The line is at writing. Pasting AI-generated prose — summaries, paraphrases, literature-review paragraphs — into your thesis or paper as your own work is where students get into genuine trouble. Most Indian universities run submissions through plagiarism-detection software, and several now run AI-content checks too (which are themselves unreliable — ironic, given that SciSpace sells its own “AI Detector” as a Premium feature; detectors flag human writing as AI-generated often enough that no detector’s verdict should be treated as proof of anything).
Practical rules that keep you safe:
- Check your university’s policy first. Some departments explicitly allow AI-assisted drafting with disclosure; others treat undisclosed AI text as misconduct. The policy — not a blog post — is what governs you.
- Use AI for discovery, write the argument yourself. Let the tools find and summarize; the synthesis, the critique, and the conclusions should be yours.
- Cite everything the tools surface. An AI-found paper still needs a real citation, verified against the original — AI tools occasionally garble references, so open the actual paper before citing it.
- Never fabricate. If a tool gives you a citation you cannot open and verify, do not use it. Hallucinated references in a thesis are a career-ending kind of embarrassment.
Who should pick what: the decision framework
You are an undergraduate or master’s student with a reading list. Start with the ₹0 stack: Elicit’s free unlimited search to find papers, SciSpace’s free 100 credits to understand the hard ones, Consensus’s free 10 Pro messages to check key claims. Pay for nothing until you hit a specific wall.
You are writing a dissertation or thesis literature review. SciSpace Premium ($12/month annual, ≈₹1,130) is the natural first purchase: unlimited chats with papers, unlimited exports to your citation manager, the paraphraser for rewording your own drafts. Add Elicit Plus ($11/month annual) if your review needs structured extraction tables.
You are a PhD scholar doing a systematic review. Elicit Pro is built for exactly this — the 5,000-paper screening workflow and 20-column extraction tables. At ≈₹44,000/year it is a lab purchase, not a personal one; ask your guide or department about institutional access before paying yourself.
You need quick evidence checks, not deep reading. Consensus Pro ($12/month annual) answers “does the evidence support this?” faster than anything else. The free tier’s 10 messages a month is enough to learn whether you actually need it.
You are a freelance academic writer or editor. This is a real income stream in India — and our AI side-hustles guide covers it. SciSpace Premium plus Elicit Plus (≈₹2,170/month combined) is a defensible business expense against client work; the exports and extraction tables are the billable features.
What the free stack can’t do
Honesty requires the other side too. None of the free tiers gives you unlimited everything, and the walls are deliberate: SciSpace blocks exports, Elicit throttles its agent, Consensus caps AI answers at 10 a month. If your research genuinely needs systematic-review scale, there is no free lunch — the paid tiers exist because the compute costs are real. The good news is that the free tiers are honest about what they are: trial ramps, not crippleware. You can do real work on all three without paying, which is more than most AI tools offer.
Sources
- Vendor pricing pages checked live October 23, 2026: SciSpace (scispace.com/pricing), Elicit (elicit.com/pricing), Consensus (consensus.app/pricing — Individual and Team/Enterprise tabs, full-text searched for INR/UPI/India mentions: none found).
- USD→INR ≈₹94.4, October 2026 (mid-market; your card is charged in USD at your bank’s rate plus forex markup).
- SciSpace/TypeSet origin: Inc42 company overview (Typeset founded 2016 by Saikiran Chandha and Shanu Kumar, headquartered Bengaluru) and Inc42 funding report, September 2017 ($850K pre-Series A); Wellfound startup listing (SciSpace, Bengaluru, 11–50 employees).
- Elicit corpus: 138M+ papers (elicit.com/pricing, October 23, 2026); built on the Semantic Scholar corpus (semanticscholar.org).
- Unverified and excluded: the “IND30” 30%-off India student code for SciSpace (secondary blog claim only — not on any official page); older reviews’ “Study Snapshots” free-tier claims for Consensus (absent from the current pricing page); Consensus student-discount eligibility for Indian institutional emails (banner exists; not testable without sign-up).
AI-assistance disclosure: this article was researched and drafted with AI assistance; every price, plan detail, and company claim above was checked against the vendor’s own site or a cited source on October 23, 2026, and hedged or excluded where the primary source was silent. Converted rupee figures use the mid-market rate — your card will be charged in USD at your bank’s rate plus any forex markup.
