Kreuzberg, Inc.
Kreuzberg (Xberg) is an ultra-fast document intelligence and RAG platform that turns any document, URL, code file, or audio clip into clean, structured data ready for AI pipelines.
At a Glance
- AI and Machine Learning teams
- Financial services for document automation
- Open-source developers
- Enterprise software companies building RAG systems
AI Tools by Kreuzberg, Inc.
(1)
Xberg
Open Source Document Intelligence SDK
Discussions
No discussions yet
Be the first to start a discussion about Kreuzberg, Inc.
Latest News
Xberg v1.0.4 released with bug fixes and full MCP server support
Core document intelligence libraries relicensed under MIT as Kreuzberg, Inc.
Announcement of Xberg v1 as successor to Kreuzberg
Kreuzberg library featured on Hacker News Show HN
Products & Services
High-performance document intelligence engine for extracting text, tables, and metadata from 101+ formats with a Rust core.
Managed extraction API with SDKs, dashboards, and observability for production workloads.
Web crawling and scraping engine supporting HTTP and headless Chrome fallback with HTML to Markdown conversion.
Fast and lossless HTML to Markdown engine used within the Xberg pipeline.
Market Position
Positions as an open-source, self-hostable, and faster alternative to cloud-only OCR services like Amazon Textract, Google Document AI, and Azure AI Document Intelligence.
Leadership
Founders
Na'aman Hirschfeld
Co-Founder & CTO at xberg.io; Creator of Litestar (formerly Starlite), Polyfactory, and basemind; high-performance polyglot open-source tools maintainer with a focus on Rust cores.
Executive Team
Na'aman Hirschfeld
Co-Founder & CTO
Creator of Litestar and basemind; background in building polyglot open-source document intelligence and code intelligence tools.
Founding Story
Built over two years as a successor to the open-source Kreuzberg library, Xberg was designed to solve the bottlenecks in AI data pipelines by providing high-performance, polyglot extraction primitives.
Business Model
Revenue Model
Managed extraction API with pay-as-you-go pricing (per page processed) and enterprise-grade support for self-hosted deployments.
Pricing Tiers
Includes a free page allowance for prototyping and small projects.
Charged in arrears for actual usage; no fixed plan fee.
Locked favorable pricing for 24 months, direct roadmap input, and early access.
Target Markets
- AI and Machine Learning teams
- Financial services for document automation
- Open-source developers
- Enterprise software companies building RAG systems
- RAG (Retrieval-Augmented Generation) Ingestion
- Document Classification and Routing
- Compliance and Document Review
- Intelligent Code Analysis