Skip to content
SPECIFICTION
Get listed

ASReview

asreview.nl

Free plan Open source WebWindowsmacOSLinuxAPISelf-hosted
Visit asreview.nl
Spec score 9.0 #1 of 38 in AI Literature Review Software

What it is

ASReview is a free, open-source tool that uses AI-aided active learning to prioritize records for systematic review screening, with a human making screening decisions. Its site says the approach can reduce screening workload by up to 95%. Users can simulate AI-aided review to compare models, while Crowdscreen supports parallel screening with access managed by administrators through a private login system. The described model families are ELAS-Ultra, ELAS-Heavy, and language-agnostic ELAS-L; ELAS-Heavy and ELAS-L require the Dory package. ASReview works with reference managers including EndNote and PubMed, and imports and exports CSV, RIS, and XLSX files. Documentation includes a Python API, while the institutional workflow also describes a REST API and CLI for automation. In a local setup, data stays on the user's computer; server deployments keep it in the organization's environment and follow its policies. Support includes documentation, tutorials, videos, community discussion, and organizational workshops and consultancy. ASReview LAB costs nothing with no licensing fees or subscriptions, though running hardware or a server may incur costs.

Spec sheet

What it does
ASReview uses active learning to reorder large collections of textual records so likely-relevant items appear earlier during systematic-review screening.Read 15 Oct 2026 · asreview.nl
Purpose
ASReview uses AI-aided active learning to prioritize records for systematic review screening while keeping a human screener in the decision loop.Read 28 Sep 2026 · asreview.nl
Intended users
The site presents ASReview for screeners, information specialists, data scientists, developers, institutions, and students.Read 28 Sep 2026 · asreview.nl
Platforms
ASReview LAB can be installed locally on macOS, Linux, and Windows, or run self-hosted on a server or in the cloud and accessed with a web browser.Read 28 Sep 2026 · asreview.nl
API
ASReview documentation includes a Python API, and the institutional page also mentions a REST API and CLI for automated workflows.Read 28 Sep 2026 · asreview.nl
Integrations
The site describes compatibility with EndNote, PubMed, and other reference managers, and says RIS files can be imported and exported.Read 28 Sep 2026 · asreview.nl
Privacy
The site says the default local setup keeps data on the user's computer; server deployments keep data within the user's environment and apply that organization's policies.Read 28 Sep 2026 · asreview.nl
Support
The site offers free online documentation, tutorials, videos, and community discussion support, and describes in-house workshops and custom consultancy for organizations.Read 28 Sep 2026 · asreview.nl
License
The ASReview software repository is released under the Apache License 2.0.Read 15 Oct 2026 · github.com
Headquarters
Utrecht, NetherlandsRead 28 Sep 2026 · asreview.nl
AI models
Version 2 presets include ELAS-ultra, ELAS-heavy and ELAS-lang, with configurable classifiers, feature extractors and query strategies.Read 15 Oct 2026 · asreview.nl
API maturity
The API documentation warns that the API is still under development and may change without warning.Read 15 Oct 2026 · asreview.readthedocs.io
Citation-manager integrations
Documented citation-manager compatibility includes EndNote, Excel, Mendeley, RefWorks and Zotero through supported export formats.Read 15 Oct 2026 · asreview.readthedocs.io
Company description
LinkedIn describes ASReview as open-source software developed at Utrecht University for active-learning screening of scientific papers, policy reports and social-media data.Read 15 Oct 2026 · nl.linkedin.com
Company headquarters
LinkedIn lists Utrecht, Netherlands as the headquarters and primary location.Read 15 Oct 2026 · nl.linkedin.com
Company size
LinkedIn lists the organization size as 11–50 employees.Read 15 Oct 2026 · nl.linkedin.com
Crowdscreen
Crowdscreen lets AI agents propose records to a crowd of experts.Read 15 Oct 2026 · asreview.nl
Custom models
Model extensions can add classifiers, query strategies, balancing strategies and feature-extraction methods.Read 15 Oct 2026 · asreview.readthedocs.io
Dataset quality requirement
Official documentation says models work best when duplicates are removed and titles and abstracts are complete for most records.Read 15 Oct 2026 · asreview.readthedocs.io
Documentation depth
The documentation covers user, server and technical guides, including authentication, command-line use, APIs and extension development.Read 15 Oct 2026 · asreview.readthedocs.io
Documentation support
The maker directs users to more than 500 pages of documentation and a GitHub community for help.Read 15 Oct 2026 · asreview.nl
Double-screening limitation
The independent review describes ASReview as a single-screener prioritization engine without multi-reviewer blind screening.Read 15 Oct 2026 · augmentedscholars.com
Extension system
Python extensions can add model, subcommand and dataset functionality through setuptools entry points.Read 15 Oct 2026 · asreview.readthedocs.io
Formats
The institutional workflow page describes CSV, RIS, and XLSX import and export.Read 28 Sep 2026 · asreview.nl
Full-text limitation
ASReview focuses on title-and-abstract screening and does not provide downstream full-text review or data-extraction workflows.Read 15 Oct 2026 · augmentedscholars.com
Human decisions
Human reviewers label records while the model continuously updates the ranking.Read 15 Oct 2026 · augmentedscholars.com
Independent praise: auditability
RSRCHDB describes the open code and peer-reviewed benchmarks as useful when screening decisions must survive audit.Read 15 Oct 2026 · rsrchdb.com
Independent praise: privacy
The Augmented Scholar highlights that local operation requires no account or upload and keeps unpublished review data on the user’s machine.Read 15 Oct 2026 · augmentedscholars.com
Independent praise: transparency
The Augmented Scholar praises the open algorithms, versioned software, peer-reviewed method and ability to benchmark simulations locally.Read 15 Oct 2026 · augmentedscholars.com
Limit: cold starts
RSRCHDB advises seeding the model with known-relevant and known-irrelevant records because cold starts are slow.Read 15 Oct 2026 · rsrchdb.com
Limit: setup complexity
RSRCHDB says installation expects comfort with Python and provides less hand-holding than managed review platforms.Read 15 Oct 2026 · rsrchdb.com
Local privacy
The default local setup runs on the user’s computer, so review data does not leave that environment.Read 15 Oct 2026 · asreview.nl
Notable event: Screenathon
The homepage reports that more than 25 European researchers screened 12,473 publications in two days during the first Screenathon.Read 15 Oct 2026 · asreview.nl
Notable news: Vici grant
The homepage reports that the project secured a €1.5 million Vici grant from NWO.Read 15 Oct 2026 · asreview.nl
Project reproducibility
ASReview project files preserve configuration details such as model hyperparameters, feature matrices and initial labels.Read 15 Oct 2026 · asreview.nl
Public Python API
The Python API can open .asreview project files and retrieve project settings, datasets, feature matrices, labels, timestamps and classifier information.Read 15 Oct 2026 · asreview.readthedocs.io
Rating on The Augmented Scholar
5 out of 5 from an editorial rating; the page does not state a review count.Read 15 Oct 2026 · augmentedscholars.com
Screening
The site says ASReview can reduce the screening workload by up to 95%.Read 28 Sep 2026 · asreview.nl
Screening workload
The maker claims screening can reduce workload by up to 95%.Read 15 Oct 2026 · asreview.nl
Search integrations
Tested search-engine exports include CINAHL, Cochrane, Embase, ERIC, OpenAlex, PsycINFO, Scopus and Web of Science.Read 15 Oct 2026 · asreview.readthedocs.io
Security claims
The site describes ASReview as GDPR compliant, says it has no tracking cookies, and says it provides security updates.Read 28 Sep 2026 · asreview.nl
Security updates
The homepage states that continuous updates are made for protection and security.Read 15 Oct 2026 · asreview.nl
Self-hosting
ASReview can run locally, on a server or cloud instance, and through Docker.Read 15 Oct 2026 · asreview.nl
Server privacy
For server deployment, the maker says data remains inside the hosting organization’s environment and is governed by its policies.Read 15 Oct 2026 · asreview.nl
Simulation
Simulation mode lets users test and compare model performance on completed reviews before using a model operationally.Read 15 Oct 2026 · asreview.nl
Stopping-rule limitation
Independent reviewers say the software cannot determine when the remaining records contain no relevant studies, so users must justify a stopping rule.Read 15 Oct 2026 · augmentedscholars.com
Supported inputs
The software accepts CSV, TSV, XLSX and RIS data containing textual records such as titles and abstracts.Read 15 Oct 2026 · asreview.nl
System requirement
ASReview LAB requires Python 3.10 or later for pip installation.Read 15 Oct 2026 · asreview.readthedocs.io
Unsupported import
PubMed exports in RIS and CSV are marked unsupported in the compatibility table.Read 15 Oct 2026 · asreview.readthedocs.io

Plans and prices

ASReview LAB Free Free and open source · no licensing fees or subscriptions · running your own hardware or server may incur costs Read 28 Sep 2026 · asreview.nl
Free and open-source ASReview LAB Free core screening software · local, server or Docker deployment · no paid tier stated Read 15 Oct 2026 · asreview.nl

Every ASReview plan and price →

Strong points and limits

Strong points

  • Human screener retains decision-making role
  • Simulation supports model comparisons
  • Crowdscreen enables parallel team screening
  • Imports and exports CSV, RIS, and XLSX
  • Local setup keeps data on the user's computer

Limits

  • ELAS-Heavy requires the Dory package
  • ELAS-L requires the Dory package
  • Running hardware or a server may incur costs
  • No AI paper summaries or full-text management

Who it suits

ASReview suits screeners, information specialists, students, and research teams handling systematic reviews. Institutions and developers may also use its team, API, or command-line workflows.

ASReview in depth

Read the full write-up

ASReview uses active learning to prioritize records in systematic-review screening while keeping a human screener in control. It suits researchers and teams who want a flexible screening workflow without licensing fees. Its strongest case is focused prioritization; look elsewhere if you need full-text management or AI paper summaries.

Overview

ASReview uses screening decisions to guide which records a person should assess next. The site says it can reduce screening workload by up to 95%; that is a potential, not a guaranteed result. It supports recommendations, systematic reviews, and collaboration, but not AI paper summaries or full-text management.

For broader comparisons, see our AI Literature Review Software, Literature Review Software, and Systematic Review Software lists.

Key features

Screening and model evaluation

Active learning prioritizes records as a human screener makes decisions, which can help focus effort without handing the screening judgment to software. Simulation lets users compare models and test AI-aided reviewing. ASReview describes ELAS-Ultra, ELAS-Heavy, and language-agnostic ELAS-L; the latter two require the Dory package, an additional dependency to account for when choosing a model.

Formats and integrations

ASReview supports CSV, RIS, and XLSX import and export, and is compatible with EndNote, PubMed, and other reference managers. These options make it easier to fit screening into an existing reference workflow. A Python API, REST API, and CLI support automated workflows, particularly useful for developers and institutions integrating screening into a broader process.

Team workflows and privacy

Crowdscreen supports parallel screening, with administrators managing access through a private login system. The default local setup keeps data on the user's computer; server deployments keep it within the organization's environment and apply its policies. That distinction gives teams a meaningful choice over where screening data resides. ASReview also describes itself as GDPR compliant, says it has no tracking cookies, and provides security updates.

Free documentation, tutorials, videos, and community discussion support are available. Organizations can also arrange in-house workshops and custom consultancy, which may help with adoption but are separate services rather than part of the no-fee software plan.

Pricing

PlanPriceWhat it includesASReview LABfreeFree and open source, with no licensing fees or subscriptions; running your own hardware or server may incur costs.There is no free trial because the plan is free. The lack of licensing charges suits students, individual screeners, and institutions alike, but self-hosting is not necessarily cost-free: organizations may need to budget for their own hardware or server. There are no published seat or screening quotas to weigh against team use.

Platforms

ASReview supports API, Linux, macOS, self-hosted, web, and Windows. Local use and server deployment offer different data-location choices, so the right setup depends partly on whether an individual computer or an organization-managed environment better fits the workflow.

Who it's for

ASReview is aimed at screeners, information specialists, data scientists, developers, institutions, and students. It is a strong fit for people conducting systematic reviews who want AI to order screening work while retaining human decisions, or for teams that need parallel screening and configurable deployment. It is a weaker fit for readers seeking one tool to manage full texts and generate paper summaries as well as screen records.

Pros and cons

Pros: Human-directed prioritization and model simulation let researchers evaluate and use AI without removing the screener from the process.

Pros: CSV, RIS, and XLSX exchange, reference-manager compatibility, and APIs support a range of existing and automated workflows.

Pros: Local and server deployment give users a choice about where data stays, while Crowdscreen supports parallel team work.

Cons: Full-text management and AI paper summaries are not included, so users needing either will need another tool or workflow.

Cons: ELAS-Heavy and ELAS-L require Dory, and running one's own hardware or server may add costs despite the absence of licensing fees.

Alternatives

Rayyan is worth considering for teams that want mobile access and a freemium option with a defined free tier of 3 active reviews and 2 free reviewers; its free tier has limits that ASReview's stated plan does not.

Universal SLR Assistant is another free, open-source option across Linux, macOS, web, and Windows; its cloud AI provider API fees may apply.

Moara may suit readers who want literature search integrations and collaboration, with its free plan capped at 50 papers per review library.

RobotReviewer is a free, open-source alternative available across API, Linux, macOS, self-hosted, web, and Windows.

Scholarcy is an alternative for readers seeking article summaries, though its free plan allows 10 summaries and one-at-a-time flashcard export.

Elicit is another freemium research option, with its Basic plan offering unlimited search across more than 138 million papers, summaries, and chat with papers, alongside limited Research Agent and Research Reports usage.

Axelium may suit someone wanting a defined free tier for one lifetime review and one seat, or a paid tier with a stated monthly price.

Silvi is an alternative with a free plan that supports an unlimited number of reviews, easy screening, and transparent data extraction.

Verdict

Choose ASReview if you need systematic-review screening that uses active learning to prioritize records while keeping people responsible for decisions, especially when local or organization-managed deployment matters. Its no-fee, open-source plan and flexible workflows are compelling; choose another tool if full-text management or AI paper summaries are central to your work.

Verdict

ASReview explains its screening workflow, model options, formats, and privacy distinctions between local and server use. The software has no licensing fee, but self-run infrastructure may have costs.

Where Specifiction places it

Alternatives to ASReviewAll alternatives →