RAG и базы знаний
private-gpt
Complete API layer for private AI apps on local models: RAG, skills, tools, MCP and text-to-SQL
★ 57,6 тыс. звёздApache-2.0
Смотреть руководство →llmware — проект с открытым исходным кодом из категории «RAG и базы знаний» от llmware-ai (llmware-ai/llmware на GitHub). У него 14 821 звезда на GitHub и 2 930 форков, он распространяется по лицензии Apache-2.0 и последний раз обновлялся на GitHub 1 октября 2026.
Unified framework for enterprise RAG pipelines with small, specialized models - local, private and secure.
Plan your pipeline around small, specialized on-device models for privacy and cost, adding a cloud model only for steps that truly need one.
Add the Python package to your environment; virtually all examples and models can run on a laptop.
Create a Library, add a folder of mixed documents (PDF, Office, text, images and more) to parse, chunk and index them, then install an embedding.
Run text, semantic or hybrid queries over the library and feed the results into a Prompt so answers are grounded in your documents, with evidence checks.
Access all models the same way through ModelCatalog, regardless of implementation - GGUF, OpenVINO, ONNX Runtime, Hugging Face, Sentence Transformers or API models.
50+ llmware fine-tuned SLIM, Bling, Dragon and Industry-BERT models (roughly 1-7B parameters) designed for RAG workflows and enterprise process-automation tasks on device.
The Library add_files function parses and chunks many formats - pdf, pptx, docx, xlsx, txt, csv, md, json, wav, png, jpg, html - routed automatically by file type.
Query libraries with text, semantic, hybrid and metadata-filtered queries, and mix and match embedding models and vector databases on the same library.
Attach files or query results directly to a Prompt as model-ready context, then run post-inference evidence checks against the sources.
Optimized for Windows, Mac and Linux across AI PC, laptop, edge and self-hosted deployments, including GPU and NPU acceleration paths.
Clone llmware for the full examples collection; most users install from PyPI instead.
git clone https://github.com/llmware-ai/llmware.git
cd llmwareInstall llmware into your Python environment.
pip install llmwareLoad a small catalog model and ask a question with context, following the README's ModelCatalog and Prompt examples.
python -c "from llmware.models import ModelCatalog; print(ModelCatalog().list_all_models())" # see README examples for the full workflowllmware is developed by llmware-ai (AI Bloks) with open-source contributors.
Yes. The framework is Apache-2.0 licensed; individual models in the catalog carry their own licenses.
No for local models: the catalog's open models run on your machine. Cloud models from OpenAI, Anthropic or Google are optional and would use your own keys.
Those are general orchestration frameworks that assume you bring models and infrastructure; llmware bundles an optimized model catalog of small local models with the full RAG pipeline in one opinionated framework.
Run pip install llmware, create a Library from a folder of documents, install an embedding, and run your first semantic query.
Complete API layer for private AI apps on local models: RAG, skills, tools, MCP and text-to-SQL
A low-code MCP framework for building complex, innovative RAG pipelines — visibly, not as a black box.
Ready-to-run templates for RAG, AI pipelines and enterprise search that stay in sync with live data
Local knowledge-base RAG and agent app built on LangChain with ChatGLM, Qwen and Llama models
Private document chat for attorneys: local LLMs, verifiable citations.
A private AI platform for agents, assistants and enterprise search over your documents
Другие проекты из категории «RAG и базы знаний» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если llmware вам не подходит.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
Соседние проекты из категории «RAG и базы знаний» в каталоге, ближайшие по числу звёзд к llmware — продолжайте изучать полку по обе стороны от него.
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
Local knowledge-base Q&A over PDFs, Office files, images and web pages - offline and CPU-friendly.
Air-gapped, open-source NotebookLM alternative that turns local documents into briefings, decks and podcasts.
The smallest vector index in the world: private RAG over everything on your laptop with 97% less storage.
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
All-in-one AI framework for semantic search, LLM orchestration and language model workflows.
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.