Xberg Unifies Document Parsing, OCR, RAG

Xberg’s appeal is consolidation: it turns fragmented ingestion pipelines into one local-first extraction layer, though real-world accuracy still needs validation against specialized parsers.

• –Supports PDFs, Office files, images, archives, audio, video, URLs, and source trees
• –Combines OCR, layout detection, table reconstruction, transcription, code intelligence, and RAG chunking
• –Rust core enables native performance, streaming, caching, WASM, CLI, REST, and MCP deployment
• –MIT licensing and bindings for 15 languages make it practical for teams building portable ingestion infrastructure
• –Its successor relationship to Kreuzberg means existing users should evaluate migration and API compatibility carefully