Kreuzberg is a document intelligence platform designed for AI engineering workflows, enabling users to extract structured, machine-readable content from a wide variety of documents. The service is aimed at integrating extracted data directly into AI agents, data pipelines, and applications, facilitating rapid and scalable document processing for tasks such as retrieval-augmented generation (RAG) and embeddings workflows.
The platform supports over 97 file formats, including PDFs (both native and scanned), images (JPG, PNG), Microsoft Office formats (DOCX, PPTX, XLSX), HTML, and plain text, with automatic detection and optimization for each type. Users can upload files via API, SDK, CLI, or Docker, and the system processes documents using layout detection, built-in OCR for scanned files, table extraction, and schema validation—all in a single API call. The output is delivered as structured JSON, with webhook options for asynchronous workflows. Kreuzberg also offers semantic chunking across code, markdown, and plain text, token reduction, keyword extraction, and rich metadata generation, making the data ready for use in AI pipelines.
For code files, the tool can extract functions, classes, imports, and symbols across 306 programming languages, providing structured outputs suitable for semantic chunking and RAG pipelines. Embeddings are generated rapidly using a Rust-native engine with four presets available out of the box and the option to extend to any model. Kreuzberg leverages vision language models as an OCR backend and can extract structured JSON from documents using schemas, supporting 146 LLM providers, including local models without requiring API key configuration.
008 per page, offering the first 10,000 pages free and no monthly minimums. High-volume discounts are available for customers processing more than 100,000 pages per month. The platform also provides a self-hosted option with the same capabilities, allowing organizations to run Kreuzberg on their own infrastructure, ensuring that no data leaves their environment. The open-source engine is also available for those who prefer to manage their own deployments. Documents are processed in memory and deleted immediately after extraction, with no data stored or used for model training.
Kreuzberg is an AI & ML product. It focuses on automating the extraction of structured data from diverse document formats for integration into AI pipelines and applications. It is built as a B2B product for AI engineers and developers building document processing or retrieval-augmented generation workflows. There is a free tier. Kreuzberg is available on the web, API, and the command line, and it can be self-hosted.
Kreuzberg first shipped in 2025. Development happens publicly on GitHub with 8.6k stars and 2.3k commits in the last 90 days. PulseGate's similarity index finds few close equivalents — Kreuzberg occupies a relatively distinct niche. Key capabilities include Document OCR, table extraction, and layout detection. It exposes integrations via a public API.
Latest indexed changes and source events
Other apps tracked under the same category.
