// rag · pgvector/pgvector

pgvector

pgvector es un proyecto de RAG y conocimiento de código abierto de pgvector (pgvector/pgvector en GitHub). Tiene 23.276 estrellas en GitHub y 1.341 bifurcaciones, se publica bajo la licencia NOASSERTION y se actualizó por última vez en GitHub el 8 de octubre de 2026.

Vector similarity search inside Postgres - store embeddings next to your data

★ 23,3k estrellas⑂ 1,3k bifurcacionesNOASSERTIONactualizado oct 2026

Ver en GitHub ↗

▦ visual: el README de este proyecto no incluye ninguna captura incrustable,
así que aquí va un marcador. La guía de abajo está completa de todos modos.

// datos del proyecto

Datos del proyecto

Repositorio: pgvector/pgvector
Estrellas en GitHub: 23.276
Bifurcaciones: 1.341
Licencia: NOASSERTION
Última actualización en GitHub: 8 de octubre de 2026
Categoría: RAG y conocimiento
Tipo de ficha: Guía de instalación completa

// cómo funciona

Cómo funciona

Decide how you will store vectors
Plan a Postgres table with a vector column sized to your embedding model's dimensions.
Get pgvector for your Postgres
Clone the repository (or use a Postgres image or managed service that ships it).
Build, install and enable it
Compile the extension, install it into Postgres, and run CREATE EXTENSION vector in your database.
Run your first similarity query
Insert a few embeddings, add an index, and query nearest neighbors with the distance operator.

// características

Por qué gusta tanto

Exact and approximate search

Run exact nearest neighbor queries or speed them up with approximate indexing as your dataset grows.

HNSW indexing

Build HNSW indexes for a strong speed-recall tradeoff on large vector collections.

IVFFlat indexing

Use IVFFlat indexes when you want faster build times and lower memory use than HNSW.

Multiple distance functions

Choose from L2, inner product, cosine, L1, Hamming and Jaccard distances per query.

Flexible vector types

Store single-precision, half-precision, binary and sparse vectors to match your memory budget.

Full Postgres power

Combine vector search with SQL filters, JOINs, transactions and any Postgres client language.

// ejecútalo

Ejecútalo tú mismo

Clone the repository
Get the source code from GitHub and enter the project directory.
```
git clone https://github.com/pgvector/pgvector.git
cd pgvector
```
Build and install the extension
Compile pgvector against your Postgres installation (the README pins the current release branch).
```
make
make install # may need sudo
```
Enable it and query
Create 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 <-> operator
```

// casos de uso

Qué puedes crear

Add semantic search to an existing Postgres app without a separate vector database
Power RAG retrieval where embeddings and business data live in one database
Build product or content recommendations with nearest-neighbor queries in SQL
Filter vector search by tenant, category or permissions using normal SQL WHERE clauses

// preguntas frecuentes

Preguntas frecuentes

Who maintains this project?

It is maintained by the pgvector open-source project, led by Andrew Kane, on GitHub.

Is it free to use?

Yes - pgvector is open source. GitHub lists its license as NOASSERTION, so review the LICENSE file in the repository for the exact terms.

Does it need a paid API key?

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.

How does it compare to alternatives?

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.

How do I get started?

Los demás proyectos de RAG y conocimiento con más estrellas del catálogo de LibreAI: la lista realista de alternativas si pgvector no es lo que buscas.

MarkItDown★ 193,5k
MarkItDown — Microsoft tool that converts PDFs, Office files, images and more to clean Markdown for LLMs, MIT
Dify★ 157,8k
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
ragflow★ 91,6k
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★ 90,8k
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
MinerU★ 82,5k
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms

// más en rag

Más en RAG y conocimiento

Proyectos de RAG y conocimiento cercanos en el catálogo, los más próximos en estrellas a pgvector: sigue explorando a ambos lados.

opendataloader-pdf★ 29,5k
Turn PDFs into AI-ready data: extract Markdown, JSON and tables — and auto-tag for accessibility
DB-GPT★ 20,1k
An agentic AI data assistant - chat with your databases, run SQL and code, get reports
FastGPT★ 29,8k
A knowledge-base platform with visual AI workflows — build Q&A systems on your own documents
DocsGPT★ 18,3k
A private AI platform for agents, assistants and enterprise search over your documents
WeKnora★ 31,9k
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
turbovec★ 17,4k
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings

Consigue pgvector

Gratis y abierto: consigue el código fuente, dale una estrella al repositorio y empieza a crear.

Ver en GitHub ↗← Más herramientas de RAG y conocimiento