Convert PDF documents to structured Markdown format with automatic image extraction and hosting. Supports URL links and base64 encoded data.

Convert images to Markdown format based on advanced visual understanding models that accurately identify text content and structural information.

High-precision text recognition focused on Chinese and English, returns results in pure text format, suitable for daily documents and screenshots.

High-precision text recognition with position detection, extracts text content and provides detailed coordinate information (rotation rectangle and four-point coordinates) for each text block.

Support for multilingual text recognition including Arabic, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish, Vietnamese and more, returning results in plain text format.

Extract structured information from invoices, certificates, and forms with custom JSON Schema support, returning results in JSON format.

Extract mathematical formulas from images and return recognition results in LaTeX format, supporting both handwritten and printed formulas.