DocTranslate: AI Document Translation That Preserves Structure, Tables and Layout
DocTranslate uses a vision-language model to deconstruct any document image or PDF into its structural components, then translates every element with a 72B LLM — preserving headings, tables, lists and two-column layouts in clean HTML output.