Turn any document into LLM-ready text.

PDF, Word (DOCX), Excel (XLSX), PowerPoint (PPTX), CSV, HTML, images (PNG/JPG/WebP, via OCR), and plain text or Markdown. Every format converts to clean Markdown, and on Pro you can extract structured JSON to your own schema.

Yes. Paste any URL and we fetch the page, strip the nav, ads, and footers, and return the main content as clean Markdown.

Conversions run on a concurrent worker pool with content-hash caching, so identical files return instantly and never re-bill. Paid plans get a dedicated fast lane that never queues behind free traffic.

You get clean, structure-preserving Markdown — headings, lists, and tables kept intact — or schema-shaped JSON, with no HTML noise. That means fewer tokens, better chunking, and more reliable retrieval for your model.

On Pro, choose JSON output and either let us infer a sensible schema automatically or paste your own. We map the document straight into typed fields — ideal for invoices, forms, and reports.

Both. Create an API key and POST a file or URL to /api/v1/convert, batch up to 25 at once via /api/v1/batch, or run live web search at /api/v1/search. Each key is rate-limited by your plan. We also ship an MCP server so agents can call us directly.

Yes — send multiple files and/or URLs in a single /api/v1/batch request and get a per-item result array back, even if some items fail.

Your files and conversions are private to your account and stored only to power history and re-conversion. You can delete them anytime, and we never sell your data. See our Privacy Policy for details.

Yes. Search the live web and get back ranked results already scraped to clean Markdown — fresh content ready for your model, from the dashboard, the API, or MCP.

Conversions are metered per month by plan, with per-minute rate limits on the API. You’ll get a clear message when you reach a limit — upgrade in one click to raise it instantly.