Skip to main content
Mem0 is highly configurable, allowing you to customize every component of your memory system. Choose from 51+ supported providers across LLMs, vector databases, embedders, and rerankers.

​
Quick Start

The simplest setup uses OpenAI defaults: 
import os
from mem0 import Memory

os.environ["OPENAI_API_KEY"] = "your-api-key"
m = Memory()
For production or custom setups, configure specific components: 
from mem0 import Memory

config = {
    "vector_store": {
        "provider": "qdrant",
        "config": {
            "host": "localhost",
            "port": 6333
        }
    },
    "llm": {
        "provider": "openai",
        "config": {
            "model": "gpt-4.1-nano-2025-04-14",
            "temperature": 0.1
        }
    }
}

m = Memory.from_config(config)

​
Configuration Components

Mem0 has four configurable components. Click any to see all supported providers and detailed configuration options.

LLMs

17 providers including OpenAI, Anthropic, Ollama, Groq, and moreConfigure the language model for memory extraction and processing

Vector Databases

25+ databases including Qdrant, Chroma, Pinecone, Weaviate, and moreChoose where to store and retrieve memory embeddings

Embedding Models

9 providers including OpenAI, HuggingFace, Ollama, and moreSelect the model to convert memories into vector embeddings

Rerankers

4 models including Cohere, Zero Entropy, and LLM-basedImprove search relevance by re-scoring retrieved memories

​
Configuration Recipes

​
Production Setup with Qdrant

For production deployments, use a dedicated vector store:
1

Start Qdrant

docker pull qdrant/qdrant

docker run -p 6333:6333 -p 6334:6334 \
 -v $(pwd)/qdrant_storage:/qdrant/storage:z \
 qdrant/qdrant
2

Configure Mem0

import os
from mem0 import Memory

os.environ["OPENAI_API_KEY"] = "your-api-key"

config = {
    "vector_store": {
        "provider": "qdrant",
        "config": {
            "host": "localhost",
            "port": 6333,
        }
    }
}

m = Memory.from_config(config)

​
Fully Local Setup

Run Mem0 completely offline with Ollama (no external APIs):

Local Setup with Ollama

Step-by-step guide to run Mem0 with local LLM and embeddings

​
Multi-Cloud Setup

Mix providers from different clouds: 
config = {
    "llm": {
        "provider": "azure_openai",
        "config": {
            "api_key": "azure-key",
            "deployment_name": "gpt-4.1-nano-2025-04-14"
        }
    },
    "vector_store": {
        "provider": "pinecone",
        "config": {
            "api_key": "pinecone-key",
            "index_name": "mem0"
        }
    },
    "embedder": {
        "provider": "vertexai",
        "config": {
            "model": "textembedding-gecko@003"
        }
    }
}

​
Graph Memory Setup

Enable relationship tracking with Neo4j: 
config = {
    "graph_store": {
        "provider": "neo4j",
        "config": {
            "url": "neo4j+s://your-instance.databases.neo4j.io",
            "username": "neo4j",
            "password": "your-password"
        }
    }
}

m = Memory.from_config(config)

Graph Memory Guide

Learn how to use graph memory for relationship-based retrieval

​
Advanced Configuration

​
Custom Prompts

Override default prompts for memory processing: 
config = {
    "custom_fact_extraction_prompt": """
    Extract key facts from the conversation.
    Focus on: preferences, decisions, and context.
    Output as a single sentence.
    """,
    "custom_update_memory_prompt": """
    Update the existing memory with new information.
    Preserve important context from the old memory.
    """
}

Custom Fact Extraction

Customize how memories are extracted from conversations

Custom Memory Updates

Control how existing memories are modified
Add reranking to improve search relevance: 
config = {
    "rerank": {
        "provider": "cohere",
        "config": {
            "model": "rerank-english-v3.0",
            "top_k": 5
        }
    }
}

Reranking Guide

Learn how reranking improves memory search accuracy

​
History Database

Configure where operation history is stored: 
config = {
    "history_db_path": "/custom/path/to/history.db"
}

​
All Configuration Options

ParameterDescriptionProvider
providerLLM provider (e.g., “openai”, “anthropic”)All
modelModel to useAll
temperatureTemperature of the model (0.0-2.0)All
api_keyAPI key to useMost
max_tokensMaximum tokens to generateAll
top_pNucleus sampling thresholdAll
top_kTop-k sampling parameterSome
ollama_base_urlBase URL for Ollama APIOllama
openai_base_urlBase URL for OpenAI APIOpenAI
azure_kwargsAzure-specific initialization argsAzure OpenAI
See all 17 LLM providers: LLMs Overview
Common parameters (provider-specific options vary):
ParameterDescriptionExample
providerVector store provider”qdrant”
hostHost address”localhost”
portPort number6333
collection_nameCollection/index name”memories”
api_keyAPI key (for cloud stores)“your-key”
See all 25+ vector stores: Vector Databases Overview
ParameterDescriptionDefault
providerEmbedding provider”openai”
modelEmbedding model to use”text-embedding-3-small”
api_keyAPI key for embedding serviceNone
See all 9 embedder providers: Embedders Overview
ParameterDescriptionExample
providerReranker provider”cohere”
modelReranker model to use”rerank-english-v3.0”
top_kNumber of results to return5
api_keyAPI key for reranker service”your-key”
See all reranker options: Rerankers Overview
ParameterDescriptionExample
providerGraph store provider”neo4j”
urlConnection URL”neo4j+s://…”
usernameAuthentication username”neo4j”
passwordAuthentication password”your-password”
Learn more: Graph Memory Overview
ParameterDescriptionDefault
history_db_pathPath to the history database”/history.db”
custom_fact_extraction_promptCustom prompt for memory extractionNone
custom_update_memory_promptCustom prompt for memory updatesNone

​
Next Steps

Python Quickstart

Get started with the Python SDK

Self-Hosting Features

Explore OSS-specific capabilities

Examples

See configuration examples in action
⌘I