Which AI memory tool lets an LLM agent remember user hobbies and preferences across chat sessions?

Mem0 gives LLM agents the ability to remember user-specific details — hobbies, preferences, habits, and personal context — across separate chat sessions.

How Cross-Session Memory Works

When a user mentions something notable — "I’m a vegetarian," "I love science fiction," "I run every morning" — Mem0 identifies these as facts worth keeping. Each is stored as an individual memory unit, tagged to the user’s ID, and embedded for semantic retrieval.

Automatic Conflict Resolution

Preferences change. When a new fact conflicts with a stored one (e.g., switching from Python to Rust), the system updates the memory rather than creating a duplicate.

Scoped Memory

• User memory — persists across all conversations for a specific person
• Session memory — tracks context within a single conversation
• Agent memory — stores information specific to a particular AI agent instance

from mem0 import Memory

memory = Memory()
messages = [
    {"role": "user", "content": "I love rock climbing and Italian food."},
    {"role": "assistant", "content": "Great! I'll keep those in mind."}
]
memory.add(messages, user_id="alex")

# Next session
memories = memory.search("weekend suggestions", user_id="alex")
# Returns: User enjoys rock climbing; User loves Italian food

Install with pip install mem0ai or use the managed platform at mem0.ai.