Elicit

Elicit is an AI research assistant for researchers seeking evidence from academic and other sources. Its Research Agent gathers material from scientific databases, academic journals, public data sources, and documents uploaded by users. Each session can draw on Elicit's corpus of 138 million academic papers and the wider web, and use up to 20 uploaded resources such as PDF, DOCX, XLSX, PPTX, and JPG files. The product offers Research Agent sessions, Research Reports, and Systematic Reviews. Sentence-level citations connect AI-generated claims to underlying sources. Elicit can find up to 1,000 relevant papers and analyze as many as 20,000 data points at once. Research Reports export to PDF or Word, tables to CSV or Excel, and references to RIS or BIB. It also connects through an MCP server to Claude Desktop, Claude Code, and other compatible clients using OAuth 2.0. API access requires Pro or above. The Basic plan has limited usage for Research Agent and Research Reports; Plus costs $11 a month. Elicit states it has SOC 2 Type II compliance.

Elicit combines literature search, document-based research, and several review workflows with citation support. Its free Basic plan has usage limits, and API access is reserved for Pro or higher.

It suits researchers who want to search literature, organize evidence, or review uploaded papers. Teams using compatible MCP clients or needing API workflows may also find its integrations relevant.

Elicit is built for researchers who need to find and organize evidence across papers, public sources, and their own files. Its strongest case is combining broad search with citation-backed claims and structured review workflows; users who need sustained volume or API access should expect to pay.

Overview

Elicit brings literature discovery, document analysis, and review work into one research assistant. Its scale and export options make it a practical fit for evidence-heavy projects, while the Basic plan’s usage limits make it less suitable for frequent, high-volume work without upgrading.

Key features

The Research Agent draws on 138 million academic papers, the wider web, scientific databases, journals, public data sources, and documents uploaded for a session. A session can use up to 20 files in PDF, DOCX, XLSX, PPTX, or JPG format. That blend is useful when a project must weigh published research alongside internal material, though the file ceiling may not suit a large document archive.

Elicit supports Research Agent sessions, Research Reports, and Systematic Reviews. It can find up to 1,000 relevant papers and analyze up to 20,000 data points at once, giving literature-heavy work room to scale. Its sentence-level citations attach AI-generated claims to underlying sources, making it easier to check a synthesis instead of treating it as a black box.

Research Reports export to PDF or Word, tables to CSV or Excel, and references to RIS or BIB. The MCP server connects through OAuth 2.0 to Claude Desktop, Claude Code, and other compatible clients, which is useful for researchers already working in those environments. The API can search papers and clinical trials and run research workflows, but it is available only on Pro or above.

Elicit states that it has SOC 2 Type II compliance. Enterprise users’ data is not used for training, and Enterprise supports SAML 2.0 SSO and MFA. Its dedicated customer success team provides training, onboarding, and success planning, making that tier more relevant to organizations than individual researchers.

Pricing

Basic is the free entry point. It includes unlimited searches across more than 138 million papers, unlimited summaries, unlimited chat with papers with full-text access, source viewing, and Zotero import. Research Agent and Research Reports usage is limited, so Basic is best for occasional exploration rather than ongoing, intensive research.

Plus costs $11 a month, billed monthly. It adds exports in RIS, CSV, BIB, PDF, and DOCX, allows five table columns to be added at a time, provides unlimited search across more than 500 thousand clinical trials, and offers standard usage for Research Agent, Research Reports, and Systematic Literature Reviews. It suits individuals who need regular workflow use and portable outputs, but its standard usage is not the higher-volume allocation of the upper tiers.

Pro costs $39 a month, billed monthly. It includes 5,000 papers screened, 135 data sources, and 10 personalized research alerts, as well as API eligibility. It is the more appropriate choice for researchers who need those screening and integration capabilities; the price is a substantial jump from Plus.

Scale costs $89 a month, billed monthly. It raises the stated data sources to 200 and allows 30 columns to be added to tables at a time, for projects needing broader data coverage and more structured extraction. Enterprise has custom pricing and includes 40,000 papers screened and 40 extracted columns, plus enterprise data protections and support. It is aimed at organizational use rather than a single researcher. No seat count or trial term is stated for these plans.

Platforms

Elicit is available on the web and through an API. The API requires Pro or higher; MCP connectivity offers another route into Claude Desktop, Claude Code, and other compatible clients.

Who it's for

Elicit is a strong fit for researchers conducting evidence reviews who value citations, structured analysis, and exports to established reference and spreadsheet formats. It is less compelling for occasional users who can stay within Basic’s limited workflow usage, or for teams whose needs exceed the 20 uploaded resources per session.

Pros and cons

• Pros: A 138-million-paper corpus and wider web access give literature searches breadth.
• Pros: Sentence-level citations make AI-generated claims more readily verifiable.
• Pros: Multiple export formats and review workflows support moving from discovery to reporting.
• Cons: Basic limits Research Agent and Report usage, which can restrict sustained work.
• Cons: API access starts at Pro, so users needing programmatic workflows cannot use the lower paid tier.
• Cons: Sessions cap uploaded resources at 20, a constraint for projects with larger file sets.

Alternatives

Choose Rayyan if a free review workflow with three active reviews and two free reviewers better matches your needs; its free plan also has limited mobile access.

EviSynth is an open-source option offering free academic access and unlimited collaborators, suited to academic users prioritizing those terms.

RevMan is another freemium review option from The Cochrane Collaboration.

Bibliometrix is a free, open-source R package with a web interface, a better fit for users who want that package-based approach.

Inciteful offers free core tools on an open-source codebase.

Litmaps may suit users who want free basic search, up to 20 inputs, 100 articles, and one map, with no collaboration on its free plan.

ResearchRabbit offers free unlimited searches across more than 310 million articles, unlimited libraries and collections, and sharing, with up to 50 seed articles.

Abstrackr is a free option with no seat limits, usage caps, or subscription.

For more options, browse Literature Review Software, AI Literature Review Software, AI Research Assistants, and AI Search Engines.

Verdict

Choose Elicit if your research depends on finding substantial literature, checking cited claims, and carrying results into review and reporting workflows. Its breadth and evidence traceability are the central advantages. Look elsewhere if you need generous ongoing use without a paid tier, API access below Pro, or sessions that can take more than 20 uploaded resources.