MEM9.AI

Unlimited memory for OpenClaw

Your agents forget everything between sessions. mem9 fixes that. Persistent memory infrastructure with hybrid search, shared spaces, and cross-agent recall from first write to forever.

Powered by TiDB Cloud TiDB Cloud

How to install

Read https://mem9.ai/SKILL.md and follow the instructions to install and configure mem9 for OpenClaw

Copy the command above into OpenClaw to get started. An API key is generated automatically — no sign-up required.

Never forget again

Cloud-persistent memory that survives resets, restarts, and machine switches.

Securely backed up

Your agent memory lives in durable cloud storage, not fragile local files.

Seamless onboarding

Start with one instruction, then bring over existing memory without breaking your flow.

Need the backend substrate behind memory? Explore via TiDB Cloud Zero →

Features

Persistent memory, zero plumbing

Stop duct-taping databases, vector stores, and sync scripts together. mem9 gives your agents one memory layer for storage, retrieval, and sharing without the wiring work.

01

Instant persistent storage

Spin up a durable memory backend in seconds. No schema design, no control plane, no ops. Your agent writes and mem9 persists.

02

Hybrid search, zero config

Keyword search works out of the box. Add embeddings and mem9 automatically upgrades to vector plus keyword with no re-indexing and no pipeline changes.

03

Memory that follows your agent

Close the tab. Restart the machine. Switch devices. Your agent's memory persists in the cloud and follows it everywhere across sessions, machines, and tools.

04

Open source, self-hostable

Apache-2.0 Go server, TypeScript plugins, and bash hooks. Run it on our cloud or bring it home. Your agent's memory, your infrastructure.

Benchmark

LoCoMo Benchmark Results

Evaluating long-conversation memory quality across multi-hop reasoning, single-hop recall, temporal reasoning, open-domain QA, and adversarial robustness.

Model: qwen3.5-plus
58.84% F1 Score
71.95% LLM Score
53.76% Evidence Recall
Category F1 Score LLM Score Evidence Recall
Multi-hop Reasoning 22.60% 53.90% 25.1%
Single-hop Recall 58.18% 76.01% 67.8%
Temporal Reasoning 13.79% 44.79% 18.6%
Open-domain QA 56.57% 79.55% 60.1%
Adversarial 96.19% N/A 57.1%

LoCoMo Benchmark — Long-Conversation Memory evaluation framework

Platforms

One memory layer. Every agent.

Agents shouldn't lose context when they switch tools. mem9 gives every agent in your stack a shared, persistent memory that stays durable, searchable, and always in sync.

OpenClaw
Unlimited memory
Give your OpenClaw agents memory that never expires. Recall past conversations, reuse learned knowledge, and stay consistent session after session.
Your Memory
Beta
Official mem9.ai app
Visualize, manage, analyze, import, and export your memories from the official mem9.ai interface.
Try Your Memory

Also works with any client that can read or write through the mem9 API layer.

Memory is often the first state problem in an agent system. When your workflow expands into files, artifacts, and retrieval, drive9 becomes the next layer.

Explore drive9 →

Security

Security & Privacy

mem9 is designed to give users the benefits of persistent cloud memory with clear operational boundaries and strong security foundations.

How mem9 handles data

mem9 stores memory data to help agents preserve useful context across sessions, devices, and workflows. The system is designed around that job: storing, retrieving, and serving memory with clear data handling boundaries around access and operations.

Core security protections

Encryption in transit and at rest

Memory data is protected while moving across the network and while stored.

Access controls

Production access is controlled and limited to the systems and operators that need it.

Auditability and operational visibility

Key actions are observable so operations can be tracked and reviewed.

Isolated data handling boundaries

Memory processing is scoped to clear service boundaries to reduce unnecessary exposure.

Production-grade cloud infrastructure / Trust foundation

The underlying platform is built for durability, reliability, and steady operations. mem9 also benefits from mature security practices, controls, and operational standards behind the scenes.

FAQ

How API keys and mem9 API work

These are the questions we receive most often from onboarding emails and first-time API users.

How do I get a mem9 API key?

There are two supported paths. The fastest path is to paste the onboarding command below into OpenClaw and let mem9 guide setup or reconnect for you.

If you want to provision a key directly, call the hosted provisioning endpoint. The response body returns an `id`, and that value is your mem9 API key.

  • Use the same key later in Your Memory or on another trusted machine.
  • Keep it private. Anyone who has the key can access that mem9 space.

Onboarding command

Read https://mem9.ai/SKILL.md and follow the instructions to install and configure mem9 for OpenClaw

Direct provisioning

curl -sX POST https://api.mem9.ai/v1alpha1/mem9s
Open SKILL.md Open API reference
Can I reuse the same API key on another machine or in Your Memory?

Yes. The same API key reconnects the same mem9 space on another trusted machine, and it is also the credential you use inside Your Memory.

If the dashboard asks for a Space ID, enter the same mem9 API key there.

Open Your Memory
How should I store the API key?

Treat the API key like a secret. Store it in a password manager, secure vault, or another controlled secret store.

Do not commit it to a repository, paste it into screenshots, or share it in public channels.

What can I do with the mem9 API?

The hosted API lets you provision a key, create and search memories, update or delete individual memories, upload memory or session files, and inspect captured session messages.

Most users only need `v1alpha2` day to day, with `X-API-Key` as the primary auth header.

Browse all endpoints
What is the difference between v1alpha1 and v1alpha2?

`v1alpha1` is mainly for provisioning a new key and for legacy tenant-scoped compatibility routes.

`v1alpha2` is the normal hosted path for reads, writes, imports, and session lookups. Send your key in `X-API-Key`, and optionally send `X-Mnemo-Agent-Id` when you want agent attribution.

See version and auth details
Is mem9 secure?

mem9 is built on enterprise-grade cloud infrastructure with encryption in transit and at rest, access controls, auditability, and clear operational boundaries.

If you need deeper trust details, start with the security overview on the homepage and reach out for additional materials.

Jump to security overview