Unstract

What is Unstract?

Open-source document processing into structured data.

Who is Unstract best for?

Unstract is best for engineering teams in finance and insurance who need structured JSON from varied documents.

• Finance and insurance paperwork

  Bank statements, claims, loan files and KYC documents from many sources can be read with one set of prompts instead of a template for each layout.

• Developers adding extraction to apps

  A finished extraction project can be deployed as a REST API, an ETL pipeline into a warehouse, an MCP server for AI agents or an n8n node.

• Teams that must keep data in house

  The open-source edition and the on-premise Enterprise plan run inside your own infrastructure, so documents never leave it.

• Fields where a wrong value is costly

  Its LLMChallenge option has a second model check each extracted value and returns nothing rather than a guess.

How much does Unstract cost?

Unstract's cloud plans cost $499 a month for Starter with 5,000 pages or $2,249 a month for Growth with 25,000 pages, both with a 14-day free trial, and the open-source edition is free to self-host.

Billed yearly, Starter is $416 a month for 60,000 pages a year and Growth is $1,874 a month for 300,000 pages a year. Enterprise runs on Unstract's cloud or your own servers, is priced on request, and adds SOC 2, HIPAA, GDPR and ISO 27001 compliance, SSO and a dedicated success manager. The open-source edition (AGPL 3.0) is free to self-host with Docker. A page is one physical page, so a 10-page PDF uses 10. LLMWhisperer, the parser included in every plan, is also sold on its own: 100 pages a day free, then $1 to $15 per 1,000 pages depending on the mode. Full pricing on unstract.com. We checked these prices on 28 Sep 2026.

How do readers rate Unstract?

No one has rated Unstract yet. Be the first.

Is Unstract open source?

Yes. Unstract's code is public under the AGPL 3.0 licence, so you can run it on your own server.

What are the best Unstract alternatives?

Unstract is an open source alternative to Airparser, Hyperscience, Affinda and Nace Document Intelligence. Interfaze 1.1 Lite does the same job and is open source too.

AirparserFree

Structured data from emails, PDFs and documents, GDPR-compliant.

HypersciencePaid

Document processing automation for enterprises.

AffindaFreemium

Data extracted from any document type, validated.

Nace Document IntelligenceFreemium

Nace's API that parses, splits, classifies and extracts 36 file types, citing each value's page.

Interfaze 1.1 LiteOpen Source

Interfaze's open-weight model for OCR, transcription and extraction, with confidence scores.

Questions about Unstract