knwler

Knwler is an open-source tool designed to transform unstructured documents into structured knowledge by automatically extracting entities, relationships, and topics. It is aimed at organizations that require large-scale document intelligence, particularly in domains such as regulatory compliance, legal analysis, research, and due diligence, where accuracy, traceability, and data privacy are essential. The platform processes documents such as PDFs, text, or Markdown, segmenting them for analysis and generating actionable knowledge graphs in a matter of minutes.

The tool features automatic schema discovery, allowing it to infer optimal entity and relation types from document samples without manual ontology engineering. Its multilingual support covers English, German, French, Spanish, and Dutch, with auto-detection and prompt adaptation for each language. Knwler is capable of fully on-premise, air-gapped operation with local large language models, ensuring that no data leaves the user's infrastructure and meeting strict data sovereignty requirements. Community detection is integrated, enabling the identification of clusters of related entities and the assignment of human-readable topic labels for thematic insights. The incremental augmentation feature allows users to process new documents and have the existing knowledge graph intelligently consolidated, rather than rebuilt from scratch.

Knwler supports export to multiple formats, including JSON, GML, GraphML, and interactive HTML, and enables direct import into platforms such as Neo4j, Gephi, yEd, Memgraph, or SurrealDB. The tool can be installed using pipx, UV, or directly from GitHub. 5 3B model, but it also supports OpenAI backends via API key configuration. The platform is released under the MIT license, and its documentation and source code are available on GitHub. Knwler is developed by Orbifold Consulting.