Mem0
ActiveDescription
Mem0 is a long-term memory layer for AI agents, supporting cross-session memory management and personalized context retrieval.
Key Features
- Multi-level memory — seamlessly manages user, session, and agent memory states with adaptive personalization
- Single-pass ADD extraction — one LLM call extracts memories with no UPDATE/DELETE, memories accumulate only
- Entity linking — entities extracted, embedded, and linked across memories for retrieval boosting
- Multi-signal retrieval — semantic search, BM25 keyword matching, and entity matching scored in parallel and fused
- Temporal reasoning — time-aware retrieval ranking for queries about current state, past events, and upcoming plans
- Cross-platform SDK — Python and Node.js SDKs with CLI, self-hosted server, and cloud platform deployment options
Use Cases
💡 Personalized AI assistants: let chatbots remember user preferences and conversation history for consistent personalized experience
💡 Customer support systems: recall past tickets and user records to provide agents with full context for fast resolution
💡 Healthcare management: track patient preferences and medical history to support personalized care
💡 AI coding assistant memory: let tools like Claude Code remember project conventions, code styles, and past decisions
💡 Adaptive learning systems: automatically adjust teaching strategies and content recommendations based on user behavior
Tags
Categories
Quick Start
# Install Mem0
pip install mem0ai
# Quick start: add and search memories
from mem0 import Memory
m = Memory()
# Add a memory
m.add("User prefers dark mode", user_id="alice")
# Search memories
results = m.search("user interface preferences", user_id="alice")
print(results)