# AnyMD — Fastest Document-to-Markdown API

> The fastest document-to-markdown converter on the market. Turn PDFs, office files, and ebooks into AI-ready Markdown in under a second — purpose-built for RAG ingestion pipelines, LLM training data prep, and automated document processing.
> **Zero data retention. We do not store, save, or inspect your documents. Only page count is recorded for billing.**

AnyMD is the fastest API for converting documents to clean, structured Markdown. Written in Rust and processing entirely in memory, it outperforms every alternative on the market — no queuing, no waiting, no disk I/O. Handles DOCX, ODT, PDF, PPTX, XLSX, RTF, EPUB, HTML, CSV, and more.

## Privacy & Data Handling

- **Zero data retention** — Uploaded files are converted in memory and discarded immediately. We never write document content to disk.
- **No document inspection** — We do not read, analyse, or save the content of your files. What you convert stays yours.
- **Page count only** — The only metric recorded per conversion is the number of pages processed. Used solely for billing and rate limiting.
- **No training on your data** — Your documents are never used to train or improve models.
- **EU-hosted** — All processing runs on European servers. Files never leave the EU.
- **No logs of content** — Server logs contain timestamps, page counts, and HTTP status codes — nothing that could reconstruct a document.

## Use Cases

- **RAG ingestion pipelines** — Convert documents to Markdown before chunking and embedding. AnyMD outputs clean, LLM-readable text that produces better vector search results than raw PDF or DOCX parsing.
- **AI training data preparation** — Turn PDF reports, office documents, and web content into structured Markdown for fine-tuning and dataset creation.
- **LLM context injection** — Prepare documents for retrieval-augmented generation (RAG) by normalizing all input formats into consistent Markdown.
- **Content migration** — Convert entire document libraries from mixed formats (PDF, DOCX, PPTX, ODT, EPUB) into a single Markdown-based CMS or knowledge base.
- **Document archiving** — Preserve documents as future-proof Markdown instead of proprietary binary formats.
- **Data extraction** — Extract tables, lists, and structure from spreadsheets, presentations, and formatted documents.

## Key Features

- **Fastest on the market** — Rust-powered, in-memory processing. No queue, no async jobs, no waiting. Most documents return in under a second.
- **In-memory conversion** — Files are never stored on disk. Process, return, done. GDPR-friendly by default.
- **Zero data retention** — We do not store or log document content. Only page count is tracked.
- **EU-hosted infrastructure** — All processing on European servers.
- **Synchronous API** — Simple request-response. No webhooks, no polling, no complexity.
- **50 MB upload limit** — Handles large documents without chunking.
- **Free tier** — 100 pages/month, no credit card required.

## Docs

- [Getting started (API docs)](https://anymd.net/en/docs): API documentation — create a key, convert your first document, endpoints, response codes, supported formats
- [Pricing](https://anymd.net/en/pricing): 10 paid tiers from 1,500 pages (€5/mo) to Unlimited, free tier with 100 pages/month
- [Privacy & data handling](https://anymd.net/en/blog/zero-data-retention-privacy): How AnyMD handles your documents — zero retention, no inspection, page count only

## API Reference

- `POST /api/convert` — Convert one file (multipart, returns `text/markdown`)
- `POST /api/convert/full` — Convert and return embedded images as base64 JSON
- `POST /api/convert/batch` — Convert multiple files
- `GET /api/usage` — Current usage and allowance
- `GET /api/history` — Recent conversions
- `GET /api/capabilities` — Supported formats and plan limits
- **OpenAPI 3.0 spec:** `https://anymd.net/openapi.yaml` — Machine-readable API specification