Databases Overview

Supported database backends for PraisonAI

Databases Overview

PraisonAI supports 22 database backends across three categories.

Installation

pip install "praisonaiagents[tools]"

Conversation Stores (6)

Store conversation history and session data.

Database	Backend Key	Docker Command
PostgreSQL	postgres	docker run -d -p 5432:5432 -e POSTGRES_PASSWORD=pass postgres:16
MySQL	mysql	docker run -d -p 3306:3306 -e MYSQL_ROOT_PASSWORD=pass mysql:8
SQLite	sqlite	No setup needed
SingleStore	singlestore	Cloud service
Supabase	supabase	Cloud service
SurrealDB	surrealdb	docker run -d -p 8000:8000 surrealdb/surrealdb

Knowledge Stores (10)

Vector databases for semantic search and RAG.

Database	Backend Key	Docker Command
Qdrant	qdrant	docker run -d -p 6333:6333 qdrant/qdrant
ChromaDB	chroma	Local file storage
Pinecone	pinecone	Cloud service
Weaviate	weaviate	docker run -d -p 8080:8080 semitechnologies/weaviate
LanceDB	lancedb	Local file storage
Milvus	milvus	docker run -d -p 19530:19530 milvusdb/milvus
PGVector	pgvector	PostgreSQL + pgvector extension
Redis Vector	redis	docker run -d -p 6379:6379 redis/redis-stack
Cassandra	cassandra	docker run -d -p 9042:9042 cassandra
ClickHouse	clickhouse	docker run -d -p 8123:8123 clickhouse/clickhouse-server

State Stores (6)

Key-value stores for session state and caching.

Database	Backend Key	Docker Command
Redis	redis	docker run -d -p 6379:6379 redis:7
MongoDB	mongodb	docker run -d -p 27017:27017 mongo
DynamoDB	dynamodb	AWS service
Firestore	firestore	GCP service
Upstash	upstash	Cloud service
Memory	memory	In-process (no persistence)

Quick Start

The simplest way to add persistence to your agent:

from praisonaiagents import Agent

# Create agent with database persistence
agent = Agent(
    name="Assistant",
    instructions="You are a helpful assistant.",
    memory={
        "db": "postgresql://localhost/praisonai",
        "session_id": "my-session"
    }
)

response = agent.start("Hello!")
print(response)

Backend Examples

PostgreSQL

from praisonaiagents import Agent

agent = Agent(
    name="Bot",
    memory={"db": "postgresql://user:pass@localhost:5432/mydb"}
)
response = agent.start("Hello!")

Next Steps

• PostgreSQL - Production conversation store
• Qdrant - Vector search
• Redis - State and caching