llama-index-storage-docstore-postgres

What it is and what it does

This package integrates Postgres as a persistent backend for LlamaIndex's document store, allowing you to store and retrieve documents outside of memory. It sits between your LlamaIndex application and a Postgres database, handling the serialization and query logic needed to persist documents across sessions.

You use it by configuring it as the docstore backend in a LlamaIndex application. It depends on llama-index-core for the core framework and llama-index-storage-kvstore-postgres for the underlying key-value storage layer that manages the actual Postgres connection and operations.

Use it for:

• Store documents indexed by LlamaIndex in a persistent Postgres database for long-running applications
• Build RAG systems that need to retain document collections across application restarts
• Share a common document store across multiple LlamaIndex instances via a shared Postgres backend
• Integrate LlamaIndex document management with existing Postgres-based infrastructure

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides a Postgres-backed document store integration for LlamaIndex, enabling persistent storage and retrieval of documents within LlamaIndex applications.

Yes, if you are building a LlamaIndex application that needs persistent document storage and already use or plan to use Postgres. Low install friction, active maintenance, MIT license, and no known vulnerabilities make it a straightforward choice for this use case. Not relevant if you don't use LlamaIndex or prefer a different storage backend.

Install

llama-index-storage-docstore-postgres on PyPI

pip install llama-index-storage-docstore-postgres

uv add llama-index-storage-docstore-postgres

poetry add llama-index-storage-docstore-postgres

Installing llama-index-storage-docstore-postgres

Before you install

Low install friction with two runtime dependencies. Active maintenance status as of March 2026. No known vulnerabilities.

License in practice

MIT license permits commercial and private use with minimal restrictions; suitable for most projects.

Quickstart

pip install llama-index-storage-docstore-postgres

from llama_index.storage.docstore.postgres import PostgresDocumentStore

docstore = PostgresDocumentStore.from_uri(connection_string="postgresql://user:password@localhost/dbname")

Requires a running Postgres instance and valid connection credentials; Python 3.10 or later.

Verify before relying

• Whether this package requires a running Postgres instance or handles connection setup automatically
• Performance characteristics and scalability limits for document storage and retrieval
• Whether it supports document indexing or only basic storage/retrieval operations

Package facts

Evidence: llama_index_storage_docstore_postgres-0.5.0-py3-none-any.whl

Tags

More Database packages