🚀 DeepSeek-OCR Demo

Convert documents to markdown, extract raw text, and locate specific content with bounding boxes. It takes 20~ sec for markdown and 3~ sec for locate task examples. Check the info at the bottom of the page for more information.

Hope this tool was helpful! If so, a quick like ❤ī¸ would mean a lot :)
Mode
Task
Examples
Input Image Mode Task Prompt

Modes

  • Gundam: 1024 base + 640 tiles with cropping - Best balance
  • Tiny: 512×512, no crop - Fastest
  • Small: 640×640, no crop - Quick
  • Base: 1024×1024, no crop - Standard
  • Large: 1280×1280, no crop - Highest quality

Tasks

  • Markdown: Convert document to structured markdown (grounding ✅)
  • Free OCR: Simple text extraction
  • Locate: Find specific things in image (grounding ✅)
  • Describe: General image description
  • Custom: Your own prompt (add <|grounding|> for boxes)