RAG・知識ベース
helix-db
An OLTP graph database with native vector and full-text search, built in Rust on object storage.
★ 6.1k スターApache-2.0
ガイドを見る →turbovecは、RyanCodraiが公開しているオープンソースのRAG・知識ベースプロジェクトです（GitHub: RyanCodrai/turbovec）。GitHubスターは17,358、フォークは1,485で、MITライセンスの下で公開されています。GitHubでの最終更新は2026年10月7日です。
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
turbovec is an in-process index - ideal when your vectors can live with your application.
Install the Python package from PyPI, or clone the repo to use the Rust API or build from source.
Build an index, add your embeddings with IDs, and persist it to a file.
Query nearest neighbors, including filtered search over approved IDs.
Quantizes vectors to a low bit rate so large indexes fit in far less memory.
Search kernels use AVX512 on x86 and NEON on ARM for high throughput.
The Rust API mirrors the Python one, with documented file formats and staged search.
Restrict results to IDs another system approved - a SQL predicate, an ACL or a time window.
Drop-in stores for LangChain, LlamaIndex, Haystack and Agno via extras such as turbovec[langchain].
The repo publishes benchmark scripts and raw results against FAISS at matched bit rates, plus a 'when to use something else' section.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/RyanCodrai/turbovec.git
cd turbovecInstall from PyPI; add a framework extra if you want the drop-in integration.
pip install turbovec
# or: pip install turbovec[langchain]The README's Python section shows creating an index, adding vectors and querying it.
# Follow the Python quick example in the README to add embeddings and searchIt is maintained by Ryan Codrai on GitHub.
Yes - turbovec is MIT licensed and free to use.
No. It is an embedded library: no server, account or API key involved.
Compared with FAISS, turbovec focuses on TurboQuant compression at low bit rates with a simpler embedded API and framework drop-ins; compared with server databases like Qdrant, it trades distribution and filtering features for zero operations.
Run pip install turbovec and follow the README's Python example to index and search your first vectors.
An OLTP graph database with native vector and full-text search, built in Rust on object storage.
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
Fast, lightweight Python embeddings from Qdrant — dense, sparse, ColBERT, image and rerankers.
Codealive-mcp, in Python — built for context engine for large codebases, exposed throug...
The smallest vector index in the world: private RAG over everything on your laptop with 97% less storage.
Graph RAG conversational knowledge base built on TiDB Serverless vector storage
LibreAIカタログに掲載されたRAG・知識ベースプロジェクトの中で最もスターの多いもの — turbovecが合わなかった場合に検討したい、現実的な候補リストです。
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
カタログの中でturbovecとスター数が近いRAG・知識ベースプロジェクト — 前後の棚もそのまま見て回れます。
A private AI platform for agents, assistants and enterprise search over your documents
Air-gapped, open-source NotebookLM alternative that turns local documents into briefings, decks and podcasts.
An agentic AI data assistant - chat with your databases, run SQL and code, get reports
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
Vector similarity search inside Postgres - store embeddings next to your data
Unified framework for enterprise RAG pipelines with small, specialized models - local, private and secure.