Skip to content
Agentmart
Submit a tool

Best MCP servers for OCR

We found 96 MCP servers for OCR in the catalog; these are the 15 that rank highest. The order comes from our rank, which weighs adoption, freshness, documentation and publisher identity. Nobody can pay for a place on this list. docling.rs, Mistral MCP and MinerU lead it. flatmark is published by the company that makes it. docling.rs has the most GitHub stars here (111). 13 of the 15 had activity in the last 30 days. Remote servers need no local install; local ones run on your machine and can reach local files. Each entry lists why it made the list and links to its install page.

Updated Oct 26, 2026 · Built from catalog data; how we rank and grade

Top 15 of 96

  1. 1
    docling.rs

    docling-project

    Documents to Markdown for agents: PDF layout, tables and OCR, Office, HTML, scans.

    • Most GitHub stars here (111)
    • Most weekly downloads here (1.3K)
    • Active in the last 30 days

    ★ 111↓ 1.3K/wkLast active Oct 24, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  2. 2

    Mistral MCP server: extract invoices and documents from text, Markdown or OCR into typed JSON.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 15Last active Sep 28, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  3. 3
    MinerU

    linxule

    MinerU document parsing API — PDFs, images, DOCX, PPTX with OCR and batch processing.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 10Last active Oct 5, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  4. 4
    PDF To Text

    Jim-Panse

    Extracts PDF text locally, with macOS PDFKit and Vision OCR fallbacks and optional AI retry.

    • Active in the last 30 days
    • Open source (AGPL-3.0)

    ↓ 110/wkLast active Oct 15, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  5. 5

    Convert PDF/CSV to Markdown with AI-powered OCR via the PromptReady cloud API.

    • Active in the last 30 days
    • Open source (MIT)

    ↓ 238/wkLast active Oct 4, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  6. 6

    Access your reMarkable tablet - read documents, browse files, extract text and OCR handwritten notes

    • Open source (MIT)

    ★ 21↓ 90/wkLast active Aug 24, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  7. 7
    Transkribus

    lazyants

    Transkribus REST and Processing APIs — collections, HTR/OCR; default v1, optional v2 beta.

    • Active in the last 30 days

    ★ 2↓ 47/wkLast active Oct 3, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  8. 8

    Extract text from local PDF documents with optional pluggable OCR.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 1Last active Oct 13, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  9. 9

    Local OCR & screen understanding for agents: read any window, find what to click.

    • Open source (MIT)

    ★ 7Last active Aug 25, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  10. 10

    MCP server for video transcripts, screenshots, and OCR on YouTube and web videos.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 1Last active Oct 12, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  11. 11
    PaperOffice AI

    paperoffice-ai

    Document AI and headless DMS: OCR, IDP, search, workflows, signatures.

    • Runs locally or as a remote server
    • Active in the last 30 days
    • Open source (MIT)

    Last active Oct 23, 2026Local (stdio), Remote (HTTP)Claude Code, VS Code, Codex, Cursor +2

  12. 12
    flatmark

    flatmark.devOfficial

    Document to Markdown MCP server: PDF, Word, PowerPoint, Excel and HTML, with OCR for large files.

    • Published by the vendor
    • Remote server, no local install
    • Active in the last 30 days

    Last active Oct 2, 2026Remote (HTTP)Claude Code, VS Code, Cursor, Windsurf

  13. 13

    Extract data, create schemas, and export results from PDFs and documents with Suparse AI OCR.

    • Active in the last 30 days
    • Open source (MIT)

    Last active Oct 19, 2026Local (stdio)Claude Code, Codex, VS Code, Cursor +2

  14. 14
    PDF Text Extractor

    fernandoguiraud16-coder

    Extract text, pages and metadata from PDFs in bulk, with OCR for scanned files.

    • Remote server, no local install
    • Active in the last 30 days
    • Open source (MIT)

    Last active Oct 21, 2026Remote (HTTP)Claude Code, VS Code, Cursor, Windsurf

  15. 15
    fm-mcp

    yrangana

    Delegate summarising, extraction, classification and OCR to the macOS 27 on-device model.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 4↓ 4/wkLast active Oct 18, 2026Local (stdio)