Product updates

Changelog

v0.6.1

Latest
minor release
  • Add a TableFormer pipeline to Local Image to Table through the Accurate option, significantly improving table recognition accuracy while keeping processing on your device
  • Add Word (.docx) downloads to Local PDF to Markdown, including edited text, tables, equations, and preserved pictures. Improve recognition accuracy with new model options and an Accurate default preset using Docling Heron INT8, CodeFormulaV2 with an INT8 decoder, and TableFormer Accurate. Recognition and export run locally in the browser

v0.6.0

major release
  • Add Local PDF to Markdown (beta). It uses layout analysis and OCR to convert PDF pages into editable text and tables, while keeping pictures, charts, and formulas as images. Processing runs entirely in your browser with WebGPU; your PDF is never uploaded
  • Add Accurate mode (TeleOCR) to Image to Table as an alternative to the default Fast mode for scanned tables and complex layouts. Images are sent to the selected recognition service
  • Add a Border-aware recognition option alongside General in Local Image to Table. It detects whether a table has borders, then uses the corresponding models to recognize its cells and structure entirely in the browser

v0.5.0

major release
  • Redesign the site with consistent tool layouts, image previews, navigation, and reading-friendly blog pages
  • Add browser-local background removal and image cropping tools
  • Enhance Image to Markdown with document layout recognition and previews for embedded images, math formulas, and Mermaid diagrams
  • Improve Image to Word exports so embedded images and math formulas display correctly in Word
  • Expand local OCR input support to WebP, GIF, BMP, AVIF, and TIFF, and add HEIC and HEIF support to Image Format Converter
  • Add English spell checking to Local Image to Text and fix table editor reinitialization
  • Refresh tool explanations, privacy information, and illustrated guides to clarify processing workflows and file handling

v0.4.5

minor release
  • Refresh Local Image to Text and Local Image to Table with consistent drag-and-drop workflows, editable results, clearer progress, and improved exports
  • Prefer WebGPU for local OCR and table recognition, with automatic WebAssembly fallback for broader browser compatibility

v0.4.4

minor release
  • Add a categorized, icon-based All Tools directory with current-page highlighting to every tool sidebar

v0.4.3

minor release
  • Promote the four primary AI tools in the main navigation, simplify the Tools menu, fix menu dismissal, and add consistent result placeholders to newer tool pages

v0.4.2

minor release
  • Upgrade Image Redactor with a FastAPI, HunyuanOCR, and Presidio pipeline, character-span bounding-box alignment, confidence metadata, and manually movable and resizable redaction regions

v0.4.1

minor release
  • Add Image Translator V2 with coordinate-based text replacement, editable translated regions, background controls, automatic text fitting, and local reconstructed PNG export

v0.4.0

major release
  • Add Image Translator with English output and Original, Translation, and Compare views
  • Add Image to Mermaid with editable source, live rendering, and SVG export
  • Add Image to Word with structured OCR review and real DOCX generation
  • Add text redaction with Hunyuan coordinate OCR, automatic common-sensitive-field matching, manual review, and local flattened PNG export

v0.3.2

minor release
  • New private Image Format Converter: export images as JPG, PNG, or WebP entirely in the browser
  • Add output quality controls and configurable transparency fill for JPG conversion

v0.3.1

minor release
  • New Image to PDF tool: convert multiple images to PDF with adjustable DPI and page size (A4, A5, Letter, etc.)
  • Add Chrome Extension page with links on the home page and footer

v0.3.0

major release
  • New Image to Formula tool: convert math photos and screenshots to LaTeX with live KaTeX preview
  • Upgrade local OCR engine from PP-OCRv5 to PP-OCRv6 via the official PaddleOCR.js SDK
  • Launch the imgtotable Blog with guides on table extraction, OCR, and LaTeX conversion
  • Fix changelog link typo in the footer

v0.2.3

minor release
  • fix reset button function when total column is less than eight

v0.2.2

minor release
  • use ai mode as index page default

v0.2.1

major release
  • fix several bugs in export functions

v0.2.0

major release
  • AI model integration for enhanced table recognition

v0.1.2

security release
  • Update Next.js to 15.5.7 to fix CVE-2025-66478

v0.1.1

minor release
  • Add sitemap.xml for SEO optimization

v0.1.0

major release
  • Initial release of imgtotable application
  • Implement table image to structured data conversion
  • Add OCR functionality with Web Workers
  • Support multiple export formats (Excel, HTML, Markdown)
  • Implement client-side processing for privacy
  • Add dark/light theme support
  • Include FAQ and privacy policy pages
Browse all tools