# DocumindHQ/documind

Open-source platform for extracting structured data from documents using AI.

Repository: https://github.com/DocumindHQ/documind
Canonical: https://ross.abutalabs.com/products/documind
Homepage: https://www.documind.xyz
Language: JavaScript
License: NOASSERTION
License Family: other
Topics: ai, document-extraction, llms, open-source, pdf-extractor, developer-tools, ocr, document-analysis, extract-data, parser, pdf, pdf-converter, pdf-extractor-llm
Last push: 2025-05-15T07:57:08+00:00

## Health v2 (maintenance only)
Score: 33/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 21, release rhythm 40, longevity 46
- inputs: {"age_days": 655, "days_push": 475, "days_rel": 558, "gap_med": 7.0, "n_releases_24m": 9}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1521, forks 63 (observed 2026-08-28T04:04:57.794569+00:00)

## What it is
Documind is an open-source Node.js library that uses AI/LLMs to extract structured JSON data from PDFs and other documents based on customizable schemas. It supports OpenAI as well as local vision models (Llava, Llama3.2-vision) and converts documents to Markdown.

## Use cases
- extract structured data from pdf documents
- parse bank statements into
- convert pdfs to markdown
- auto-generate extraction schemas from documents
- run document extraction with local llms
- extract fields from invoices or receipts with a custom schema

## When to choose
- you need schema-driven structured JSON output from unstructured documents
- you want an npm library that works with OpenAI or local vision models
- you need multi-format support (PDF, DOCX, PNG, JPG, TXT, HTML)

## When to avoid
- you need a fully managed hosted extraction API without self-setup
- you require a non-AGPL license for commercial embedding
- you need high-throughput OCR without LLM costs

## Facets
- artifact type: library
- maturity: active
- function: ocr, parser, pdf, llm-inference, nlp, machine-learning
- domain: artificial-intelligence, large-language-models, pdf, files, developer-tools
- platform: cross-platform, cli
- tags: document-extraction, structured-data, pdf-to-, schema-driven-extraction, llm, openai, local-llm, markdown-conversion, natural-language-processing, nodejs

## Member repositories
- DocumindHQ/documind (main) score 33

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:57.794569+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:31:49.944460+00:00, confidence not recorded.
  - readme: https://github.com/DocumindHQ/documind (fetched 2026-08-28T04:04:57.794569+00:00, sha 0224b699fc22)
  - homepage: https://www.documind.xyz (fetched 2026-08-29T11:35:03.047388+00:00, sha 6f672c440333)
  - registry_npm: https://registry.npmjs.org/documind (fetched 2026-08-29T11:35:03.057475+00:00, sha 89c6e590681f)
- Data as of 2026-08-30T08:39:29.467469+00:00.