State that's still there tomorrow.in development
An agent that forgets everything when the session ends starts every job from zero. Memory will let state outlive the run — what an agent learned, ready the next time it wakes up. It's a piece of the agent runtime we're building now, on top of the layer that's already live.
Knowledge that survives the session.
Today an agent's memory is whatever you remembered to save and reload yourself. Memory will make that the runtime's job — durable, per-agent state that's simply there when the next run begins, with nothing to wire on your side.
State that outlives the session
An agent that stops on Tuesday will start on Wednesday knowing what it knew. Memory will hold across runs — the session can end without the knowledge ending with it.
What it learned, not just what it said
Not a chat transcript. The durable things — facts it established, decisions it made, where it got to — will persist as memory the next run can actually build on.
Scoped to the agent
Memory will belong to an agent, kept apart per agent and per key. One worker's context won't bleed into another's; what each one knows stays its own.
There on the next run
No vector store to stand up, no retrieval to wire, no state file to thread through by hand. The agent will pick up where it left off — recall handled by the runtime, on the key you already hold.
Part of the runtime. On proven ground.
Memory is one piece of the agent runtime — Layer 02 of the platform — which stands on the inference layer serving traffic in production today. When your agents start remembering, they'll be remembering on ground that's already carrying real work.
Start on the inference layer today — your key will carry over.
Your agent doesn't change.
Everything underneath does.
Start free · no card · Starter from $5/mo