AI Agent Hub
Back to skills
Free PDF Translation with Original Layout icon

Free PDF Translation with Original Layout

Office Efficiency Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_e263a450/translate-pdf according to https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Translating PDF papers directly often breaks equations, figures, tables of contents, and annotations, while a translation-only file lacks side-by-side context.

How It Works

  • Uses pdf2zh-next for parseable PDFs and preserves equations, figures, TOC, and annotations.
  • Generates xxx.zh-CN.mono.pdf, xxx.zh-CN.dual.pdf, and xxx.zh-CN.glossary.csv by default.
  • Defaults to the free SiliconFlow engine; switch to --openai or --deepseek if quality is insufficient.
  • Supports --pages, --lang-in/--lang-out, --output, --qps, and --gui.

Boundaries

  • Best for academic or well-structured PDFs; JSON warnings from the free engine usually trigger automatic retries.
  • Large files may use over 1.3GB peak memory, so translate page ranges in batches.
  • The installed binary may not be in PATH, so invoke it with the full path.

Use Cases

  • When reading English papers, translate selected pages to Chinese while keeping equations and figures
  • When comparing source and translation, generate a bilingual PDF and review the glossary
  • When localizing Chinese documents into English, translate a PDF from Chinese to English
  • When large papers cause high memory use, translate selected page ranges in batches

Best For

  • Graduate students reading English papers who need layout-preserving translation and side-by-side review
  • Technical specialists localizing foreign patents who need batched page translation and preserved layout
  • Instructors preparing course materials who need bilingual PDFs and glossary output
  • Engineers writing technical reports who need Chinese PDFs translated into English