Open Source Memory Layer

Give Your AI Perfect Memory

Add persistent, intelligent memory to any AI application. Extract facts from conversations, search semantically, and build knowledge graphs.

# Add a memory
curl -X POST https://api.memoid.dev/v1/memories \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"messages": [{"role": "user", "content": "I love hiking"}]}'

# Search memories
curl -X POST https://api.memoid.dev/v1/search \
  -d '{"query": "What are the user hobbies?"}'

Everything You Need

A complete memory layer for AI applications. Simple to integrate, powerful in production.

Persistent Memory

Store and retrieve facts from conversations. Your AI remembers users across sessions.

Semantic Search

Find relevant memories using natural language. No exact matches needed.

Knowledge Graph

Extract entities and relationships. Build rich knowledge bases automatically.

Real-time Sync

Memories update instantly. Always have the latest context available.

Privacy First

Your data stays yours. Self-host or use our secure cloud.

Simple API

Three endpoints to get started. SDKs for Python, JavaScript, and Go.

Built For

From chatbots to enterprise applications, Memoid powers memory for any use case.

AI Assistants

Build assistants that remember user preferences, past conversations, and personal details.

Customer Support

Give support agents instant context about customer history and previous issues.

Personalization

Deliver personalized experiences based on learned user behavior and preferences.

Simple Integration

Get started in minutes. Just three API calls to add memory to your application.

  • Add memories from conversations
  • Search with natural language
  • Build knowledge graphs automatically
  • SDKs for Python, JavaScript, Go
from memoid import MemoryClient

client = MemoryClient("your-api-key")

# Add memory from conversation
client.add(
    messages=[
        {"role": "user", "content": "I'm vegetarian"},
        {"role": "assistant", "content": "Got it!"}
    ],
    user_id="user_123"
)

# Search memories
results = client.search(
    query="What dietary restrictions?",
    user_id="user_123"
)

print(results[0].memory)
# "User is vegetarian"

Ready to Build AI with Memory?

Start free. No credit card required. Upgrade when you're ready.