LlamaIndex
An MIT-licensed Python framework for document-based AI apps. It works with Ollama, while its separate document services run locally or in the cloud.
LlamaIndex is an MIT-licensed Python framework for developers building AI agents and apps that answer questions using their own data. It connects documents and other sources to language models, then helps an app find the relevant material when a user asks something. Its open source framework can work with models served through Ollama.
The framework can draw from PDFs, documents, APIs and SQL databases. Developers can organize that material in indexes or graphs and use retrieval tools to supply context to an LLM. It covers the path from source data to an answer, with room to customize how information is selected and passed to a model. The README also describes integrations with LangChain and Flask.
LlamaIndex has separate tools for document parsing. LiteParse processes PDFs, Office documents and images locally, without sending them to a cloud service or using LLM tokens. It produces text and bounding box output, making it a fit for teams that need a local document parser before building a search or AI workflow.
LlamaParse handles harder documents through agentic OCR. It can read handwriting, tables and charts, extract data into a defined structure, split documents into sections, and classify them using natural language rules. It also supports indexing for retrieval and document agents that carry out multi-step work. LlamaParse requires an account and runs in LlamaIndex's cloud or can be deployed in a customer's VPC. That gives teams a choice between the local LiteParse parser and a document platform built for complex layouts and extraction.
