Compare

Three products say “agent memory.” We solve a different one.

Mem0 remembers your user. Zep knows what’s true now. Lore keeps your agent team in sync — with consistency guarantees, access control, and a token bill that goes down.

Three products, three problems

“Agent memory” means at least three different things — here’s who solves which.

Personalization memory

Mem0

Remembers your user

Durable preferences and facts about a person, ranked at read time.

Temporal memory

Zep

Knows what’s true now

Every fact carries a validity window, so stale truth is superseded, not confusing.

You’re here

Coordination memory

Lore

Keeps your team in sync

Read-your-writes, per-agent access control, and a token bill that goes down.

Side by side

Not everything each tool does — the dimensions that actually decide it for a team of agents.

Capability	Mem0	Zep	Lore for agent teams
The problem it solves
Focus	User personalization	Temporal truth	Team coordination
Consistency for a team of agents
Read-your-writes contract	–	–	✓seq + covered_seq
Working memory (hot lane)	–	–	✓synchronous CAS
Governance
Per-agent access control	basic scopes	governed messaging	✓SQL + quarantine
Efficiency & openness
Deterministic packs + savings meter	–	–	✓pack_hash
OSS scope	engine	library only	full server, 1 binary

Competitor cells reflect public docs — capability parity. If your problem is personalization or temporal truth, Mem0 or Zep may fit better.

How we compare

Category trust is part of the product — the standard every comparison here holds to.

Same judge, or it doesn’t count

Every score is reproduced with the same harness — so you can run it on your own workload, not take our word for it.

We’ll tell you when they win

If your problem is really personalization or single-assistant temporal truth, we say so — and point you to Mem0 or Zep.

Numbers and contracts, no adjectives

No “blazing-fast,” no “most advanced.” The API contract and the matrix do the talking.