# neuml/txtai

💡 All-in-one AI framework for semantic search, LLM orchestration and language model workflows

Repository: https://github.com/neuml/txtai
Canonical: https://ross.abutalabs.com/products/txtai
Homepage: https://neuml.github.io/txtai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: python, search, nlp, semantic-search, vector-search, txtai, llm, vector-database, language-model, transformers, sentence-embeddings, large-language-models, information-retrieval, search-engine, embeddings, retrieval-augmented-generation, rag, ai, agents, ai-agents
Last push: 2026-08-26T12:30:37+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 95, longevity 100
- inputs: {"age_days": 2215, "days_push": 7, "days_rel": 34, "gap_med": 26, "n_releases_24m": 26}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12909, forks 883 (observed 2026-08-28T04:11:00.071268+00:00)

## What it is
txtai is an all-in-one open-source AI framework for semantic search, LLM orchestration and language model workflows, built around an embeddings database combining vector indexes, graph networks and relational databases. It supports RAG pipelines, autonomous agents, multimodal indexing and exposes web and MCP APIs with bindings for JavaScript, Java, Rust and Go.

## Use cases
- build semantic search over my documents
- create a RAG chatbot over a knowledge base
- index and search text, PDFs, audio and images with embeddings
- orchestrate LLM workflows and pipelines
- build autonomous AI agents that search a knowledge base
- run question-answering and summarization pipelines locally
- set up a self-hosted vector database with SQL support

## When to choose
- you want an all-in-one Python framework combining vector search, RAG, pipelines and agents
- you need multimodal embeddings (text, documents, audio, images, video) with SQL querying
- you want batteries-included defaults to prototype AI search quickly
- you need MCP or web APIs with multi-language bindings

## When to avoid
- you need a standalone production vector database at massive scale (consider dedicated vector stores)
- you only need a thin embedding library or raw transformer access
- your stack is not Python-centric and you don't want to run a service

## Facets
- artifact type: framework
- maturity: active
- function: vector-database, search-engine, rag, agent-framework, llm-inference, nlp, machine-learning
- domain: artificial-intelligence, large-language-models, machine-learning
- platform: python, self-hosted, cross-platform
- tags: semantic-search, embeddings-database, llm-orchestration, vector-search, mcp, workflows, question-answering, retrieval-augmented-generation, search, natural-language-processing, ai-agents, docker

## Member repositories
- neuml/txtai (main) score 98

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:00.071268+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:13:41.756996+00:00, confidence not recorded.
  - readme: https://github.com/neuml/txtai (fetched 2026-08-28T04:11:00.071268+00:00, sha f39e3d522481)
  - homepage: https://neuml.github.io/txtai (fetched 2026-08-29T08:09:41.763166+00:00, sha 2bc1eaad271e)
  - registry_pypi: https://pypi.org/pypi/txtai/json (fetched 2026-08-29T08:09:41.768772+00:00, sha 9153ee817cf4)
- Data as of 2026-08-30T08:39:29.467469+00:00.