Elicit

Overview

Elicit is an AI research assistant for people who need to find and synthesize evidence from academic literature and other sources. It suits researchers whose work depends on cited claims, structured review workflows, and exportable results. Its breadth and citation support are compelling, but meaningful research volume and API access push users toward paid plans.

Key features

The Research Agent draws on a corpus of 138 million academic papers, the wider web, public data sources, and documents uploaded for a session. Each session can use up to 20 files in PDF, DOCX, XLSX, PPTX, or JPG format. That combination makes Elicit useful when a project needs to bring published research together with a team’s own materials; the 20-file ceiling may constrain sessions built around larger document collections.

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, a substantial scale for literature-heavy work. AI-generated claims include sentence-level citations to their underlying sources, which helps readers check the evidence rather than accept a summary on trust.

Research Reports export to PDF or Word, tables to CSV or Excel, and references to RIS or BIB. The MCP server connects to Claude Desktop, Claude Code, and other compatible clients through OAuth 2.0. API access supports paper and clinical-trial searches and research workflows, but requires Pro or above, so it is not an option for Basic or Plus users.

Pricing

Elicit is freemium. Basic has no stated price and pairs limited Research Agent and Research Reports usage with unlimited search across more than 138 million papers, unlimited summaries, unlimited full-text paper chat, source viewing, and Zotero import. It is a useful starting point for discovery and reading, but the limits on the central research workflows make it a less dependable fit for sustained synthesis.

Plus costs $11 a month, billed monthly. It adds exports in RIS, CSV, BIB, PDF, and DOCX formats, allows five table columns to be added at a time, and includes unlimited search across more than 500,000 clinical trials. Research Agent, Research Reports, and Systematic Literature Review usage is standard rather than unlimited. This is the entry paid plan for researchers who need exports and regular but bounded workflow use.

Pro costs $39 a month, billed monthly. It includes 5,000 papers screened, 135 data sources, and 10 personalized research alerts. Pro is the minimum plan for API access, so it is the relevant step up for users integrating Elicit’s searches or workflows into compatible systems.

Scale costs $89 a month, billed monthly, and raises data sources to 200 while allowing 30 table columns at a time. It suits larger or more structured analysis needs, though the plan details given do not establish a higher screening allowance than Pro. Enterprise has custom pricing and includes 40,000 papers screened and 40 extracted columns; its dedicated customer success team provides training, onboarding, and success planning.

Platforms

Elicit is available on the web and through an API. Its MCP server also connects to Claude Desktop, Claude Code, and other MCP-compatible clients using OAuth 2.0, while direct API access requires Pro or above.

Who it's for

Elicit is best for researchers conducting evidence-heavy work who value traceable claims, large-scale paper discovery, and structured research outputs. Basic can serve readers focused on search and paper chat; Plus is more suitable when exports and recurring workflows matter. Teams handling enterprise data may value the stated SOC 2 Type II compliance, SAML 2.0 SSO, MFA, and commitment not to train on Enterprise user data. Readers seeking a mobile app or unrestricted Research Agent use on the cheapest plan should look elsewhere.

Pros and cons

• Pro: Sentence-level citations make AI-generated claims easier to verify against source material.
• Pro: Paper discovery and analysis scale to 1,000 papers and 20,000 data points at once.
• Pro: Research outputs and references export in formats suited to documents, spreadsheets, and reference managers.
• Con: Basic limits Research Agent and Report usage, despite its broad paper search and unlimited summaries.
• Con: API access starts at Pro, a sizable monthly commitment for users who need only programmatic access.
• Con: The listed platform support is web and API, not a native mobile app.

Alternatives

For a broader shortlist, see AI Research Assistants, Literature Review Software, and AI Search Engines.

• Consensus is another freemium research option, with Android, iOS, web, extension, and API platforms; its free plan includes limited Pro messages and monthly allowances for Deep reviews, Study Snapshots, and API/MCP calls.
• Scholarly is a web-based alternative with a free plan and trial, though its free tier is tightly capped across chat, file uploads, and study tools.
• SciSpace offers web, iOS, and extension access; choose it if those platforms fit better, while noting that its Basic plan uses 100 monthly credits.
• Perplexity is available across web, mobile, desktop, and API platforms, making it an option for readers who want broader device coverage; its free use is limited daily.
• GPT Researcher is worth considering for self-hosting and avoiding product-level usage limits, provided you are prepared to pay underlying model and search-provider costs directly.
• Kimi Deep Research offers broad platform coverage and a free monthly allowance measured by token consumption rather than query count.
• Gemini Notebook has web, iOS, and Android access; its free plan allows 50 sources and uploads up to 200 MB per file.
• Scite is another freemium option, with a Connect tier offering MCP access and grounding in more than 300 million papers but no Assistant or Search access.

Verdict

Choose Elicit if your research depends on cited evidence, structured literature workflows, and exports that carry findings into other tools. Its paper scale and source-linked claims are the clearest reasons to pick it. Look elsewhere if you need native mobile access, or if limited Basic usage and the Pro threshold for API access do not fit your budget or workflow.