Docling MCP

What it does

Convert PDFs and other documents to structured formats via Docling, for AI applications.

Description as published by the publisher in the official MCP Registry, read 10 Oct 2026. Not written or verified by Few Servers.

Facts

Install

• PyPI package, as published: docling-mcp · version 3.3.0 · transport stdio

  uvx docling-mcp

  Standard run command for PyPI packages. The registry publishes the package name, not a command; check the repository for required arguments.

Install details read from the registry record on 12 Oct 2026. We have not run this server.

Keys and settings

The registry record declares a secret value (a key or token) that is not marked required. Declared by the publisher:

• DOCLING_MCP_CONVERSION_MODE (environment variable, not marked required): Conversion mode: 'remote' (default) or 'local' (requires docling-mcp[local] extra).
• DOCLING_MCP_SERVICE_URL (environment variable, not marked required): Base URL of the Docling Serve instance. Required when conversion mode is remote.
• DOCLING_MCP_SERVICE_API_KEY (environment variable, secret, not marked required): API key for authenticating with Docling Serve.
• DOCLING_MCP_SERVICE_TIMEOUT (environment variable, not marked required): Request timeout in seconds for the remote Docling Serve API. Default: 300.0.
• DOCLING_MCP_SERVICE_MAX_RETRIES (environment variable, not marked required): Maximum number of retry attempts for remote API requests. Default: 3.
• DOCLING_MCP_FALLBACK_TO_LOCAL (environment variable, not marked required): Set to 'true' to fall back to local conversion when Docling Serve is unreachable.
• DOCLING_MCP_KEEP_IMAGES (environment variable, not marked required): Set to 'true' to retain page images in converted documents. Default: false.
• DOCLING_MCP_IMAGES_SCALE (environment variable, not marked required): Image scale factor (e.g. 1.0, 2.0). Increase to avoid tensor batching errors. Default: 1.0.
• DOCLING_MCP_DO_OCR (environment variable, not marked required): Set to 'false' to disable the OCR pipeline. Default: true.
• DOCLING_MCP_DO_TABLE_STRUCTURE (environment variable, not marked required): Set to 'false' to disable table structure detection. Default: true.
• DOCLING_MCP_DO_PICTURE_DESCRIPTION (environment variable, not marked required): Set to 'true' to run a vision-language model to generate textual descriptions for figures and pictures. Requires a VLM to be available on the conversion backend
• DOCLING_MCP_PICTURE_DESCRIPTION_PRESET (environment variable, not marked required): Preset ID for the picture-description VLM (e.g. 'smolvlm', 'granite_vision'). When unset the backend default is used. Only applies when DOCLING_MCP_DO_PICTURE_D
• DOCLING_MCP_IMAGE_EXPORT_MODE (environment variable, not marked required): Controls how images are rendered when exporting to Markdown. Accepted values: 'placeholder' (default, emits <!-- image -->), 'embedded' (base64 data-URI), 'refe
• DOCLING_MCP_CACHE_MAX_DOCUMENTS (environment variable, not marked required): Maximum number of documents to keep in the in-memory cache. Oldest (least-recently-used) documents are evicted when the limit is reached. Default: 10.

Links

RepositoryRegistry record (JSON)PyPI package

Being listed is not a safety check. Read the repository and check what a server asks for before you connect it. How the directory works.