Elicit
AI research assistant that searches papers and builds cited reports and tables
Linked from elicit.com · checked 29 Oct
- Tool type
- AI research assistant
- Made for
- Researchers and evidence teams
- Runs on
- Web app, plus API and MCP
- API access starts on Pro; MCP server announced
What it does
Elicit takes a research question, uploaded documents or a set of papers and returns cited outputs. It searches over 138 million academic papers and 545,000 clinical trials, then produces research reports, tables of extracted data, figures, calculations and slides. Claims in its answers are backed by sentence-level citations to the underlying sources.
Researchers use it to run semantic searches without guessing keywords, summarise and chat with papers, and build customised reports in which they can change which papers and information are covered. A dedicated systematic review workflow automates screening and data extraction, while Routines let the Research Agent keep a topic current and report back for review. A Library stores sources for reuse and is described as a shared hub for people and AI agents. It goes beyond chat with interactive tables and multi-step workflows.
It suits academics, pharmaceutical and medical-device teams, policy analysts and other evidence-focused professionals. Using it requires a free account; paid plans raise usage limits and add exports, larger table and screening capacity, Routines, API access, collaboration and admin controls. Free use of the Research Agent and Reports is limited, and the larger systematic-review capacity needs Pro or above. Outputs still need expert checking, particularly for high-stakes decisions.
What it can do
- Semantic search across 138 million papers and 545,000 clinical trials
- Customisable cited research reports
- Systematic review screening and data extraction
- Research Agent producing tables, figures, calculations and slides
- Routines that monitor new research and update work
- Library for storing and sharing sources
Works with
- Zotero (import)
- Elicit API
- MCP server
Example workflows
- Literature report on a question
- Input: a research question. Steps: run a search, adjust which papers and data fields are included, generate the report. Output: a structured report with sentence-level citations.
- Systematic review screening and extraction
- Input: a set of candidate papers and inclusion criteria. Steps: screen papers, define extraction columns, review extracted data. Output: a table of extracted data points with source citations.
Before you choose
- Research Agent and Reports usage is limited on the free plan; Elicit uses a monthly usage pool rather than workflow-based limits
- Systematic review capacity and API access depend on plan tier
- Accuracy and time-saving claims come from Elicit and its customers, so verify outputs for important decisions
Listed under
Get started
Free Basic plan; paid plans upgrade usage
- Create an account via the sign-up page
- Start on the free Basic plan or choose a paid plan
- Import sources from Zotero or upload papers to your Library
Plans and prices
Prices in USDPrices are per user per month when billed annually. Monthly-billed prices were not confirmed. Enterprise is custom with sales contact.
- Unlimited paper search and summaries
- Limited Research Agent and Reports usage
- Standard Research Agent usage
- Exports and clinical trial search
- 5x usage and systematic review workflow
- Routines and API access
Source: elicit.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
Ratings and reviews
No ratings for Elicit yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- elicit.com: product, pricing and policy pages
- YC directory mirror
- Batch and company name
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.