Xberg

Extract clean text, tables, and structured data from documents and code — no format detection, no OCR setup, no stitched-together libraries. One engine, 15 language bindings, runs anywhere.

Xberg is the next iteration of Kreuzberg. Same document-intelligence engine, rebuilt and rebranded under a fresh v1 line.

Feed documents → get clean text, tables, metadata, transcripts, code intelligence · Run it library, CLI, REST API, or MCP server · No GPU needed · Stream multi-GB files · Cache results.

Documents · Images · Spreadsheets · Email · Archives · Code · Audio · Video

Quick start · What you get · Capabilities · CLI · Docs

---

---

What you get

Point Xberg at anything — a PDF, a spreadsheet, a scanned image, an audio file, a source tree — and get back clean, structured content you can use right away. One core does the format detection, reading, and extraction, so you don't assemble a pipeline yourself. Call it from Rust, Python, Node.js, Go, Java, C#, Ruby, PHP, Elixir, Dart, Swift, Zig, WASM, Kotlin, or C FFI, and run it as a library, CLI tool, REST API, or MCP server.

What it does	How
Extract from 97 formats	PDFs, Office, images, HTML, email, archives, scientific publications, and code — intelligent MIME detection, streaming for large files.
6 output formats	Plain text, Markdown, Djot, HTML, JSON tree structure, or Structured (JSON with OCR metadata and bounding boxes).
Code intelligence	Functions, classes, imports, symbols, docstrings from 306 programming languages. Syntax-aware chunking for RAG pipelines.
Crawl & recurse	Follow URLs, extract documents from within documents (nested archives, embedded PDFs). Auto/Document/Crawl modes.
OCR on demand	Tesseract, PaddleOCR, Candle, or VLM backends — fallback chains, extensible via plugins. Confidence scores. Language auto-detection.
Transcription	Whisper ONNX for audio/video tracks (MP3, M4A, WAV, WebM, MP4).
Embeddings & search	Local (ONNX models) or provider-hosted (OpenAI, Anthropic, Google, 143 providers via liter-llm). Reranking.
Structured outputs	LLM-powered extraction — local (Ollama, LM Studio, vLLM) or remote (OpenAI, Anthropic, Google).
Enrichment	NER, redaction, summarization, translation, QR code detection, page classification, keyword extraction (YAKE/RAKE), language detection, layout detection, table extraction, token reduction (TOON).
Batch & parallel	Process 100s of documents in parallel. Per-file timeouts. Configurable batch concurrency (max_concurrent_extractions).
Caching	Content-hash cache keys — skip re-extraction when the file and config are unchanged.
Deployment	Library, CLI (12 commands), REST API (xberg serve), MCP server (9 tools, 3 prompts, 4 resources), Docker.

---

Installation

Language Packages

pip install xberg

See Python README for full documentation.

npm install @xberg-io/xberg

See Node.js README for full documentation.

cargo add xberg

See Rust README for full documentation.

go get github.com/xberg-io/xberg

See Go README for full documentation.

Available on Maven Central as io.xberg:xberg. See Java README for the dependency snippet.

dotnet add package Xberg

See C# README for full documentation.

gem install xberg

See Ruby README for full documentation.

composer require xberg-io/xberg

See PHP README for full documentation.

Add {:xberg, "~> 1.0"} to your mix.exs dependencies. See Elixir README for full documentation.

npm install @xberg-io/xberg-wasm

See WebAssembly README for full documentation.

Available on Maven Central as io.xberg:xberg-android. See Kotlin README for the dependency snippet.

Add via Swift Package Manager. See Swift README for full documentation.

dart pub add xberg

See Dart README for full documentation.

Add via zig fetch. See Zig README for full documentation.

Build from source as part of this workspace. See C (FFI) README for full documentation.

CLI & Deployment

brew install xberg-io/tap/xberg

12 commands: extract, batch, detect, formats, version, cache (stats/clear/manifest/warm), serve, mcp, api, embed, chunk, completions.

See CLI usage guide for detailed documentation.

docker pull ghcr.io/xberg-io/xberg:latest

Run in API, CLI, or MCP modes. See Docker guide for examples.

xberg serve --host 0.0.0.0 --port 8000

One POST endpoint handles all formats. Returns JSON or Markdown. Stream large files. See API server guide.

xberg mcp --transport stdio

9 tools (extract, extract_batch, detect_mime_type, cache_stats, list_formats, cache_clear, get_version, cache_manifest, cache_warm). 3 prompts (extract_document, ex