# NanoNets/docstrange

Extract and convert data from any document, images, pdfs, word doc, ppt or URL into multiple formats (Markdown, JSON, CSV, HTML) with intelligent structured data extraction and advanced OCR.

Repository: https://github.com/NanoNets/docstrange
Canonical: https://ross.abutalabs.com/products/docstrange
Homepage: https://docstrange.nanonets.com/
Language: Python
License: MIT
License Family: permissive
Topics: image-to-markdown, llm, markdown, ocr, pdf-to-markdown, structured-data, ai, document-parser, document-parsing, pdf-parser, pdf-to-json, structured-data-capture, tables
Last push: 2025-10-31T14:00:01+00:00

## Health v2 (maintenance only)
Score: 40/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 49, release rhythm 35, longevity 28
- inputs: {"age_days": 398, "days_push": 306, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1531, forks 136 (observed 2026-08-28T04:04:58.794915+00:00)

## What it is
DocStrange is a Python library and tool that converts documents (PDF, DOCX, PPTX, XLSX, images, URLs) into Markdown, JSON, CSV, or HTML using advanced OCR and a 7B vision-language model. It supports both a free cloud API and fully local processing, includes a built-in web UI, and offers an MCP server for Claude Desktop integration.

## Use cases
- convert pdf to markdown for llm ingestion
- extract tables from scanned pdfs into
- ocr images and turn text into structured data
- parse word and powerpoint documents into markdown
- extract specific fields from documents given a  schema
- prepare rag-ready clean text from messy documents
- batch convert documents offline without sending data to the cloud

## When to choose
- you need high-accuracy document-to-markdown conversion with table extraction
- you want a choice between free cloud processing and 100% private local mode
- you need structured extraction against a JSON schema from mixed document types
- you want an alternative to docling, marker, or markitdown with OCR built in

## When to avoid
- you only need simple text extraction from plain-text or well-structured HTML files
- you cannot run a 7B model locally and cannot use cloud processing for privacy reasons
- you need a lightweight pure-Python parser with no ML model dependencies

## Facets
- artifact type: library
- maturity: active
- function: ocr, pdf, parser, nlp, machine-learning, llm-inference, rag, data-science
- domain: pdf, files, artificial-intelligence, developer-tools
- platform: python, windows, cli
- tags: document-conversion, document-parsing, pdf-to-markdown, structured-data-extraction, table-extraction, document-ai, mcp-server, web-ui, cloud-api, local-processing, natural-language-processing, data-engineering, linux, macos, web-server, gpu

## Member repositories
- NanoNets/docstrange (main) score 40

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:58.794915+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 2026-08-30T04:31:21.646878+00:00, confidence not recorded.
  - readme: https://github.com/NanoNets/docstrange (fetched 2026-08-28T04:04:58.794915+00:00, sha 23a83853e051)
  - homepage: https://docstrange.nanonets.com/ (fetched 2026-08-29T11:33:37.803439+00:00, sha 3b474da30d6c)
  - registry_pypi: https://pypi.org/pypi/docstrange/json (fetched 2026-08-29T11:33:37.806101+00:00, sha 4dddab6c78de)
- Data as of 2026-08-30T08:39:29.467469+00:00.