Consensus
Consensus answers research questions with study references. Explore Deep Reviews, the Consensus Meter, evidence checks, subscriptions and privacy.
Provider sources checkedWhat does Consensus do?
Who is this tool for?
- Students and specialists exploring a research area.
- Analysts and editorial teams checking claims against scholarly evidence.
Strengths
- Research questions lead directly to publications and evidence.
- Deeper analysis and collections support repeatable exploratory research.
- The meter and comparison tables expose differing positions.
Limitations and requirements
- Selected results do not establish complete coverage of all studies.
- A cited source may address a different population or question.
- Citation counts and journal indicators do not replace study-design assessment.
Who is behind the product?
- Company
- Consensus NLP, Inc.
- Headquarters
- San Francisco, United States, according to the company page.
- Company founded
- Founded in 2021 by Christian Salem and Eric Olson.
- Product launch
- The provider dates the public search-engine launch to September 2022.
The company's founding and search-engine launch occurred in different years. The current product adds analysis and integration capabilities.
Suggested ways to use it
Suggested workflows or verification steps. These are not hands-on tests; for unverified products they are not a recommendation to use the service.
Suggested workflow: turn a claim into a testable question
- Define the claim's population, period and outcome explicitly.
- Run a search and, if needed, a Deep Review; record counterevidence separately.
- Read the key original studies and phrase the conclusion only as strongly as supported.
Goal: An evidenced fact check with visible uncertainty, not a GatwAI hands-on test.
Suggested workflow: prepare a literature-review starting point
- Record search terms and criteria before creating a topical collection.
- Ask for a table of important papers, methods and unresolved questions.
- Supplement the selection with a specialist database and document exclusion reasons.
Goal: An auditable starting list for your own review, not a completed systematic synthesis.
Pricing, allowances and data
What should you check before choosing a plan?
September 2026 plan help lists free unlimited paper searches, ten Pro messages and three Deep Reviews monthly. Pro costs US$20 monthly or $144 annually; Deep costs $65 monthly or $540 annually. API/MCP allowances and deeper research have separate limits; check institutional and team arrangements separately.
What should you check about data and permissions?
The November 1, 2026 policy excludes model training on user content. Content is still sent to contracted model providers to produce results; brief retention and safety checks may apply. Uploads are generally account-private unless users share them. Confidential materials still require organizational approval.
External reviews and available evidence
A May 2026 JMIR study found no fully correct five-question sets for Consensus across 15 retracted papers.
Reported positives
- Original research; no conflicts declared.
Reported cautions
- It tested free access in 2025, not the current version.
These reports reflect the linked authors’ experiences. Test date, plan and task affect results; they do not establish a representative overall rating.
Product development and notable changes
Citations and references in the API
The changelog adds Get Citations and Get References in beta, following full-text access for API and MCP the previous day.
Source for this change ↗
Which alternative fits which task?
Frequently asked questions
Does the meter establish definitive scientific consensus?
It displays the distribution of selected findings. Quality, applicability and completeness require additional assessment.
Can I search papers for free?
Yes. Advanced analysis and integrations have separate allowances; they are not equivalent to unlimited paper search.
