Python library and SDK that builds a hierarchical tree index of documents so an LLM can reason through it for retrieval, without vector databases or chunking.
Python library that uses LLMs and graph-based pipelines to extract structured data from websites and local documents from a natural-language prompt.