RAG и базы знаний
seekdb
seekdb — the AI-native search database unifying vector, text and structured data for agents.
★ 3,1 тыс. звёздApache-2.0
Смотреть руководство →pgvector — проект с открытым исходным кодом из категории «RAG и базы знаний» от pgvector (pgvector/pgvector на GitHub). У него 23 276 звёзд на GitHub и 1 341 форк, он распространяется по лицензии NOASSERTION и последний раз обновлялся на GitHub 8 октября 2026.
Vector similarity search inside Postgres - store embeddings next to your data
Plan a Postgres table with a vector column sized to your embedding model's dimensions.
Clone the repository (or use a Postgres image or managed service that ships it).
Compile the extension, install it into Postgres, and run CREATE EXTENSION vector in your database.
Insert a few embeddings, add an index, and query nearest neighbors with the distance operator.
Run exact nearest neighbor queries or speed them up with approximate indexing as your dataset grows.
Build HNSW indexes for a strong speed-recall tradeoff on large vector collections.
Use IVFFlat indexes when you want faster build times and lower memory use than HNSW.
Choose from L2, inner product, cosine, L1, Hamming and Jaccard distances per query.
Store single-precision, half-precision, binary and sparse vectors to match your memory budget.
Combine vector search with SQL filters, JOINs, transactions and any Postgres client language.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/pgvector/pgvector.git
cd pgvectorCompile pgvector against your Postgres installation (the README pins the current release branch).
make
make install # may need sudoCreate the extension in your database, then store and search vectors with plain SQL.
CREATE EXTENSION vector;
-- then create a table with a vector column and query with the <-> operatorIt is maintained by the pgvector open-source project, led by Andrew Kane, on GitHub.
Yes - pgvector is open source. GitHub lists its license as NOASSERTION, so review the LICENSE file in the repository for the exact terms.
No. pgvector itself needs no API key; you only need a way to generate embeddings, which can be a local model or a provider API.
Unlike standalone vector databases such as Qdrant or Milvus, pgvector keeps vectors inside Postgres, so you get transactions, JOINs and one less system to operate - a good default when you already run Postgres.
Clone the repo, build and install the extension, run CREATE EXTENSION vector, then follow the README's Getting Started section to store and query your first vectors.
seekdb — the AI-native search database unifying vector, text and structured data for agents.
Deep Lake — a multimodal datalake and vector store for AI, with version control and training streaming.
Vespa — the AI search platform: vector, text and tensor search at production scale.
Meta's library for efficient similarity search and clustering of dense vectors, in C++ and Python
An OLTP graph database with native vector and full-text search, built in Rust on object storage.
Federated AI search and RAG across 100+ apps — answers with sources, without moving your data
Другие проекты из категории «RAG и базы знаний» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если pgvector 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 pgvector — продолжайте изучать полку по обе стороны от него.
Turn PDFs into AI-ready data: extract Markdown, JSON and tables — and auto-tag for accessibility
An agentic AI data assistant - chat with your databases, run SQL and code, get reports
A knowledge-base platform with visual AI workflows — build Q&A systems on your own documents
A private AI platform for agents, assistants and enterprise search over your documents
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.