AI-Powered Document Translation with Layout Preservation
Upload any PDF or image, run structured OCR, translate with DeepSeek AI, and export in your preferred format. Tables, stamps, watermarks, and multi-column layouts are preserved throughout the pipeline.
A full-suite AI document translation platform. Upload PDF or images, run structured OCR with Baidu PaddleOCR-VL, translate with DeepSeek AI, refine in the workbench editor, and export to PDF, DOCX, HTML, or Markdown. Preserves tables, stamps, watermarks, and multi-column layouts.
Upload a file first. Start processing is enabled after upload.
Your file is transmitted over encrypted connections. It is kept for 7 days for download, then permanently deleted from our servers.
Files are isolated by account. Only you can access your own files.
How It Works
Upload Your Document
Drop a PDF or image file. Supports PDF, PNG, JPG, JPEG, and BMP — up to 50MB per file.
AI Processes & Translates
Structured OCR extracts text with layout. DeepSeek AI translates with context awareness.
Edit & Export
Refine in the workbench editor, then export to PDF, Word, HTML, or Markdown.
Frequently Asked Questions
How does AI PDF translation preserve my layout?
Our structured OCR engine parses each page into text blocks, tables, and images with spatial coordinates. Translation runs on this structured representation, keeping the document hierarchy intact. The workbench editor lets you fine-tune positioning before export.
What file formats are supported for translation?
Upload PDF, PNG, JPG, JPEG, or BMP files up to 50MB each. For multi-page documents, PDF is recommended. Images are handled as single-page documents with the same full pipeline support.
Can I translate without changing the target language?
Yes. Select 'No translation (OCR only)' as the target language. The pipeline will still run OCR and make the recognized text available in the workbench editor. You can edit and export the original text without any translation applied.
How many credits does translation consume?
Each page (PDF page or single image) costs 10 credits. New users receive 30 free Mana Points upon sign-up. Credits cover the full pipeline: OCR, AI translation, workbench editing, and all export formats.
What's the difference between PDF Translator and Image Translator?
PDF Translator is optimized for multi-page PDF documents with complex layouts, tables, and academic formatting. Image Translator handles single images — screenshots, photos, scanned pages — with the same OCR and translation pipeline. Both tools share the same workbench and export options.
Explore More Document Tools
