ASReview

What is ASReview?

ASReview is open-source software for screening large sets of records in systematic reviews and other text-screening projects. You load a dataset, label a few records as relevant or irrelevant, and the model re-ranks the rest so likely matches appear first, which lets reviewers stop earlier than a full read-through. The site says this can reduce screening workload by 95%. The product has four parts: Screen, AI, Simulate for testing models on labelled data, and Crowdscreen, where AI agents propose records to a group of experts who work in parallel. ASReview LAB runs on your own computer by default, or on a server or in Docker for teams. It is coordinated at Utrecht University in the Netherlands, has no tracking cookies, and reads and writes CSV, RIS and XLSX files so it fits next to reference managers. The institutions page states there are no licence fees or subscriptions, only the cost of any hardware or server you run yourself.

Use cases

• Screen titles and abstracts for a systematic review
• Prioritise the records most likely to be relevant
• Simulate AI-aided screening on a labelled dataset
• Screen records in parallel with a team of experts
• Compare screening models before starting a review
• Run screening on a shared server for a team

Who it's for

• Systematic reviewers
• Information specialists
• Data scientists
• University researchers
• Students

Why use ASReview

• Spend reviewer time on the records that matter, since the model keeps the likely matches near the top of the queue.
• Keep sensitive data in your own environment, because the default install runs locally and the site lists no tracking cookies.
• Check how well the AI would work on your own topic before committing, using the simulation tool on a labelled set.
• Finish faster as a team with crowd screening, where several experts work through AI-proposed records at once.
• Pay nothing in licence fees, with documentation, learning materials and a weekly community session included.

ASReview features

• Active-learning screening that re-ranks records as you label them
• Simulation mode to test and compare model performance on labelled data
• Crowdscreen mode where AI agents propose records to a group of experts
• Local install for macOS, Linux and Windows, plus server, cloud and Docker setups
• Import and export of CSV, RIS and XLSX files
• Documented Python and REST APIs and a command-line interface
• Logged actions and settings for reproducible reviews
• Extra tools: Datasets, Datatools, DORY models, Insights, Makita and Templates
• Remote login through your web server or middleware for institutions

ASReview integrations

• Docker
• Zotero

How to get started with ASReview

1. Install ASReview LAB on your computer, on a server or cloud account, or with Docker.
2. Import your search results as a CSV, RIS or XLSX file and create a new project.
3. Label a few records as relevant and irrelevant so the model has examples to learn from.
4. Keep screening: the model retrains after your decisions and shows the records it thinks are most relevant next.
5. Stop when you judge enough relevant records are found, then export the labelled dataset for the next step.

ASReview limitations

• Runs on your own computer or server, so there is no vendor-hosted version to sign up for
• Advanced simulation and custom models need data-science skills and a Docker or CLI setup
• Server deployments carry your own hardware or hosting costs
• The pricing URL on the site returns a 404, so cost information comes from the institutions page

ASReview pricing

ASReview is open source, so you can use and self-host it for free.

ASReview plans

• ASReview LAB

  Free

  open source, no licence fees

  • Screen, AI, Simulate and Crowdscreen
  • Local, server, cloud or Docker install
  • Documentation and open learning materials

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

ASReview alternatives

The main ASReview alternatives listed on Debuly are Rayyan, Elicit, Synthesa, BitBank.nz, ChatPDF and DDScore. The table compares pricing and platforms; order is A–Z.

1. 

   BitBank.nz

   AI crypto price forecasts for 70+ pairs, a simulated trading bot with published fills, and a REST and WebSocket API for developers.

   Research and searchDeveloper platformsAutomation and agents
   FreemiumVisit BitBank.nz (official site)
2. 

   ChatPDF

   Upload a PDF, Word or PowerPoint file and ask questions in any language, with answers linked to the source page.

   Research and searchProductivityWriting
   FreemiumVisit ChatPDF (official site)
3. 

   DDScore

   AI first-pass due diligence for private companies: upload a deck or data room and get 12-dimension scores, exit scenarios and risks.

   Research and searchProductivity
   Visit DDScore (official site)
4. 

   Elicit

   AI research assistant that searches 138 million papers, extracts data into tables and runs systematic reviews and cited reports.

   Research and searchProductivityWriting
   FreemiumVisit Elicit (official site)
5. 

   Rayyan

   Web and mobile platform for systematic and literature reviews, with deduplication, screening, relevance ranking and team collaboration.

   Research and searchProductivity
   FreemiumVisit Rayyan (official site)
6. 

   Synthesa

   AI-assisted evidence synthesis service for systematic reviews, abstract screening and meta-analysis.

   Research and search
   Contact salesVisit Synthesa (official site)

ASReview FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from asreview.nl · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.