kreuzberg

kreuzberg

Multilingual document intelligence framework that extracts text, metadata, and structured information from various file formats.

MCPDataOpen source文档处理OCR音频转录代码智能
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 8.5k
Source
mcp-github

Overview

Kreuzberg is a multilingual document intelligence framework built on a Rust core. It can extract text, metadata, tables, and code intelligence from 96 file formats including PDFs, Office documents, and images. Supports native bindings for 16 programming languages and can run as a library, CLI, REST API, or MCP server. Ideal for modern AI and RAG pipelines requiring fast and reliable information extraction.

Capabilities

  • Extract text from 96 file formats
  • Extract metadata and structured information
  • Code intelligence for 306 programming languages
  • Built-in OCR and audio transcription capabilities
  • Supports streaming processing of large files

Use cases

Document content extractionCode intelligence analysisOCR and audio transcriptionData preprocessing in AI and RAG pipelines

Setup

Requires: API KeyNode 环境
Install via npm: `npm install @kreuzberg/node` or use the Docker image: `docker pull ghcr.io/xberg-io/kreuzberg`

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

What file formats does Kreuzberg support?

Supports 96 file formats, including PDFs, Office documents, images, and more.

Does Kreuzberg support OCR?

Yes, it includes built-in OCR engines such as Tesseract, PaddleOCR, and EasyOCR.

Related skills