MarkItDown Document Intelligence
Production-grade document conversion service with OCR, Vision intelligence, cross-page table merging, schema extraction, and audio/video transcription built on Microsoft's MarkItDown.
About
Extends Microsoft's MarkItDown with hybrid routing that selects the optimal processing path per document type. Scanned PDFs route through Mistral OCR-3 with LLM correction and dual-pass Vision validation. Text PDFs use pdfplumber with cross-page table merging. Embedded images in DOCX and PPTX are classified and described instead of dropped. Supports document classification with confidence scoring, structured field extraction against any JSON Schema, audio and video transcription via faster-whisper, and quality scoring on every response. Runs as both an MCP server (SSE + stdio) and a REST API in a single Docker container.
Is this your product? Claim this listing with an email at your domain to post updates.
Is this your project?
Claim this listing to manage your page, access analytics, and unlock upgrades. Verification takes 60 seconds.
Share This Project
Embed Badge
Add this badge to your README:
[](https://hifriendbot.com/ai-list/markitdown-document-intelligence/)
