LangExtract: Python Library for Text Data Extraction

LangExtract is a Python library designed for extracting structured information from unstructured text documents using large language models (LLMs). It effectively processes various materials, such as clinical notes and reports, by identifying and organizing key details while ensuring that the extracted data corresponds accurately to the source text.

Key features:

• Precise Source Grounding - Maps every extraction to its exact location in the source text, enabling visual highlighting for easy traceability and verification.
• Reliable Structured Outputs - Enforces a consistent output schema based on user-defined examples, leveraging controlled generation in supported models like Gemini.
• Optimized for Long Documents - Uses an optimized strategy of text chunking and parallel processing to enhance extraction accuracy.
• Interactive Visualization - Generates a self-contained, interactive HTML file to visualize and review extracted entities in their original context.
• Flexible LLM Support - Supports cloud-based models as well as local open-source models through the built-in Ollama interface.

LangExtract is adaptable to any domain, allowing users to define extraction tasks with minimal examples, thus eliminating the need for extensive model fine-tuning.