txtai

txtai is an all-in-one AI framework designed for semantic search, LLM orchestration, and language model workflows. At its core is an embeddings database that combines vector indexes (both sparse and dense), graph networks, and relational databases. This foundation enables powerful vector search and serves as a knowledge source for LLM applications. The framework supports building autonomous agents, retrieval augmented generation (RAG) processes, and multi-model workflows. Key capabilities include vector search with SQL, topic modeling, graph analysis, and multimodal indexing. It can create embeddings for text, documents, audio, images, and video. Pipelines handle LLM prompts, question-answering, labeling, transcription, translation, and summarization. Workflows allow joining pipelines together, and agents intelligently connect embeddings, pipelines, workflows, and other agents. txtai provides web and Model Context Protocol (MCP) APIs with bindings for JavaScript, Java, Rust, and Go. It is built with Python 3.10+, Hugging Face Transformers, Sentence Transformers, and FastAPI, and is open-source under an Apache 2.0 license. The framework is designed to run locally or scale out with container orchestration.