Welcome to ContextGem Documentation!#
ContextGem is a free, open-source framework that makes it radically easier to extract structured data and insights from documents using LLMs and decision models — with minimal code.
Note
New in 1.0: decision models. Next to LLMs, ContextGem now works with decision models, which answer closed questions (yes/no, one of your labels, a level of a scale) with the probability of each possible answer, fast and at a low cost.
Upgrading from 0.x? Some classes and parameters are renamed or removed, some defaults have changed, and Python 3.10 is no longer supported. Follow the migration guide.
Learn about the motivation behind ContextGem and how it works.
Instructions to install ContextGem and quickly start using it.
Learn how to create and configure documents to extract information (aspects and concepts) from.
Learn how to identify and extract specific document sections like clauses, chapters, or terms using ContextGem’s Aspects API.
Learn how to extract and infer structured data like JSON objects, strings, numbers, dates, booleans, ratings, and labels from documents using ContextGem’s Concepts API.
Learn how to create reusable extraction pipelines that combine aspects and concepts for consistent document analysis across multiple files.
Learn about supported cloud LLM providers and local models, and how to configure and use them for extraction.
Learn how to extract yes/no, label and rating concepts, as well as aspects, with decision models, which return probabilities instead of generating text.
Learn how to use ContextGem in agentic workflows: at the ingestion of documents, in a tool that an agent calls, and at the decision points of a workflow.
Explore advanced features and techniques for extracting data from documents.
Learn how to optimize your extraction pipeline for accuracy, cost, and performance.
Learn how to serialize and deserialize ContextGem objects for storage and transfer.
Complete API documentation for all ContextGem modules and classes.