RAG e conhecimento
infigraph
AST-powered code intelligence engine — graph database plus hybrid semantic search across 62 languages, zero LLM dependency.
★ 91 estrelasNOASSERTION
Ver o guia →LibreAI / RAG e conhecimento / txtai
// rag · neuml/txtai
txtai
txtai é um projeto de RAG e conhecimento de código aberto de neuml (neuml/txtai no GitHub). Tem 13.000 estrelas no GitHub e 922 forks, é publicado sob a licença Apache-2.0 e foi atualizado pela última vez no GitHub em 8 de outubro de 2026.
All-in-one AI framework for semantic search, LLM orchestration and language model workflows.
★ 13,0 mil estrelas⑂ 922 forksApache-2.0atualizado out 2026
▦ visual: o README deste projeto não inclui nenhuma captura incorporável,
então aqui vai um marcador. O guia abaixo está completo mesmo assim.
então aqui vai um marcador. O guia abaixo está completo mesmo assim.
// dados do projeto
Dados do projeto
- Repositório
- neuml/txtai
- Estrelas no GitHub
- 13.000
- Forks
- 922
- Licença
- Apache-2.0
- Última atualização no GitHub
- 8 de outubro de 2026
- Categoria
- RAG e conhecimento
- Tipo de ficha
- Guia de instalação completo
- Site oficial
- Official site
// como funciona
Como funciona
- Start with search or a workflow
Decide whether your first job is semantic search over content, a RAG pipeline, or a language-model workflow such as summarization or transcription.
- Install txtai
Run pip install txtai (Python 3.11+); install only the optional dependencies your components need to keep the footprint low.
- Index your content
Create an Embeddings instance and index text, documents, audio, images or video - sensible defaults get you running in a couple of lines.
- Build up to pipelines and agents
Chain pipelines into workflows, ground an LLM with your embeddings for RAG, or connect components into an agent; optionally serve it all through the built-in API.
// recursos
Por que as pessoas gostam tanto
Embeddings database
A union of sparse and dense vector indexes, graph networks and a relational database, enabling vector search with SQL, topic modeling, graph analysis and multimodal indexing.
Multimodal embeddings
Create embeddings for text, documents, audio, images and video, so mixed content lives in one searchable index.
Language-model pipelines
Ready pipelines run LLM prompts, question-answering, labeling, transcription, translation, summarization and more.
Workflows
Join pipelines together and aggregate business logic - from simple microservices to multi-model workflows.
Agents
Agents built on smolagents connect embeddings, pipelines, workflows and other agents, supporting Hugging Face, llama.cpp and LiteLLM-backed LLMs (OpenAI, Claude, Bedrock).
API and language bindings
A built-in FastAPI web API and MCP API, with bindings for JavaScript, Java, Rust and Go, so applications in any language can use a txtai service.
Advertisement
// execute você mesmo
Execute você mesmo
- Clone the repository
Clone txtai for the examples and source; most users install from PyPI instead.
git clone https://github.com/neuml/txtai.git cd txtai - Install from PyPI
Install txtai into a Python 3.11+ virtual environment.
pip install txtai - Serve it as an API
Point the built-in API at a YAML config and query it over HTTP.
CONFIG=app.yml uvicorn "txtai.api:app" # then: curl "http://localhost:8000/search?query=positive"
// casos de uso
O que você pode criar
- Building semantic search over a document or product catalog that understands meaning, not just keywords.
- Creating a RAG chatbot grounded in your own knowledge base, with citations.
- Running a workflow that transcribes audio, summarizes it and labels topics in one chain.
- Indexing images and text into the same space for cross-modal similarity search.
Advertisement
// perguntas frequentes
Perguntas frequentes
Who maintains txtai?
txtai is developed by NeuML, the company behind the project, with open-source contributors; NeuML also offers consulting and a hosted txtai.cloud.
Is txtai free to use?
Yes. It is open source under the Apache-2.0 license.
Does it need a paid API key?
Not for local models: txtai runs Hugging Face and llama.cpp models on your machine. Cloud LLMs reached through LiteLLM (OpenAI, Claude, AWS Bedrock) use your own provider keys.
How does txtai compare with LangChain or a standalone vector database?
A vector database only stores and searches embeddings, and LangChain orchestrates external components; txtai bundles the embeddings database, pipelines, workflows and agents into one framework with batteries included.
How do I get started?
Run pip install txtai, create an Embeddings index in a few lines, and work through the introductory notebook.
// continue explorando
Ferramentas semelhantes
RAG e conhecimento
trieve
All-in-one API platform for search, recommendations, RAG and analytics — self-hostable
★ 2,7 mil estrelasMIT
Ver o guia →RAG e conhecimento
Dify
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
★ 157,8 mil estrelasCustom (check repo)
Ver o guia →RAG e conhecimento
unstract
Unstract — LLM-driven extraction of unstructured data, built for API deployments and ETL pipeline workflows.
★ 7,3 mil estrelasAGPL-3.0
Ver o guia →RAG e conhecimento
pipeshub-ai
The open-source, permission-aware context layer that lets agents search all your company knowledge.
★ 3,8 mil estrelasApache-2.0
Ver o guia →RAG e conhecimento
vexor
Semantic search engine designed for files and code.
★ 242 estrelasMIT
Ver o guia →// alternativas
Alternativas em RAG e conhecimento
Os outros projetos de RAG e conhecimento com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se txtai não for o ideal para você.
- MarkItDown★ 193,5 mil
MarkItDown — Microsoft tool that converts PDFs, Office files, images and more to clean Markdown for LLMs, MIT
- Dify★ 157,8 mil
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
- ragflow★ 91,6 mil
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
- PaddleOCR★ 90,8 mil
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
- MinerU★ 82,5 mil
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
// mais em rag
Mais em RAG e conhecimento
Projetos de RAG e conhecimento próximos no catálogo, os mais próximos em estrelas de txtai: continue explorando dos dois lados dele.
- LEANN★ 13,0 mil
The smallest vector index in the world: private RAG over everything on your laptop with 97% less storage.
- cocoindex★ 11,7 mil
Incremental engine that keeps RAG indexes and agent context fresh by processing only the delta.
- QAnything★ 14,2 mil
Local knowledge-base Q&A over PDFs, Office files, images and web pages - offline and CPU-friendly.
- xberg★ 9,4 mil
Polyglot document intelligence in Rust — extract text, tables and structured data from 106 formats.
- llmware★ 14,8 mil
Unified framework for enterprise RAG pipelines with small, specialized models - local, private and secure.
- paper-qa★ 9,3 mil
PaperQA2 — high-accuracy RAG for answering questions from scientific papers, with citations you can check.
Baixe o txtai
Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.