Kemory.
One memory across every AI.
- MCP-native
- Permissioned & auditable
- 87.8% LongMemEval-500*
*Internally benchmarked on LongMemEval-500 using open-source models.
Six AIs. Six Memories.
None of Them Yours.
ChatGPT saves memories. Claude remembers a project. Manus carries knowledge between runs. Each one remembers you, but only on its own — what you tell one is invisible to the next, and moving it means exporting a file and pasting it in by hand. You’re not carrying one memory between tools. You’re maintaining several partial copies of your knowledge, and none of them agree.
What each one holds is thin, too. They’re good at learning your tone and habits — far less reliable at surfacing the one decision you made eight weeks ago, out of thousands, at the moment it actually matters. And none of it is governed in a way an organisation can accept. You can’t prove which memories exist, who reached them, or that they’re truly deleted.
Memory has to outlive the session, the model swap, the vendor. It has to retrieve precisely, and it has to be permissioned and auditable.
That’s not a feature. It’s infrastructure. That’s the gap Kemory is built to close.
Keep every AI chat
in one memory.
A single permissioned memory every AI and agent can share — MCP-native, auditable, and yours to hold.
One memory
every AI shares.
Stop managing memory. Start using it.
Every AI you talk to has amnesia, and a wall around it. You repeat yourself to each one, and none of them share what they learn. Kemory is a single, living memory every AI and agent can draw on — and a permission engine that decides who sees what. Shared where you want it, sealed where you do not.
No more “remember this” — Kemory remembers on its own, quietly, across every AI. What matters is kept; what you delete is gone. Compaction runs in the background on open models, so every AI receives just what it needs, already distilled.
- connect over MCPClaude Code · Cursor
- it rememberscompacted
- continue in another AIpermissioned
- or ask Koraauditable
Save tokens
and time.
Compression of your knowledge, not accumulation, is the real flex.
Every AI conversation gets more expensive the longer it runs — and not in a way you’d notice, because the cost is buried inside your subscriptions. Cost-per-turn climbs each iteration and by turn fifty you’re consuming exponentially, because the model keeps re-summarising its own history just to keep going.
It’s like using a Ferrari to go to the grocery store just for milk, every single turn — your premium AIs are re-reading what you already said, before answering what you actually asked.
Kemory inverts that curve. Instead of accumulating raw history, it compresses continuously — reducing raw memories to namespace summaries, and those into a single, synthesised context. Compaction runs on an open-weight model, so the frontier model only ever sees the compacted version — not the pile.
Memory in one AI
is a walled garden.
Kora for Chrome sits quietly in your browser and brings every AI conversation into one memory you own.
You already think across ChatGPT, Claude, Gemini and half a dozen tabs — and everything you work out in one vanishes to the others. Kora for Chrome ends that. As you chat, it gathers the conversations, files and ideas you’re creating and weaves them into your Kemory — automatically, in the background, with nothing to remember to do.
- Works where you already work — ChatGPT, Claude, Gemini and more, no copy-paste, no exports.
- Yours, not ours — local-first, syncing only to the Kemory you connect. You hold the keys, and you can see and delete anything.
- On from the moment you install — no setup ritual; open a conversation and it’s already remembering.
Ask many models,
get one answer.
One memory,
in every tab.
Everything you’ve ever worked out — on hand wherever you’re working, added to your prompt in one click.
Gathering is only half the job. Kora for Chrome brings your memory back to you, everywhere you type. The hardest part of prompting was never the asking — it’s the explaining: the background, the decisions already made, the thing you tried last month. Enhance with Kora does the explaining for you: one click in any AIs prompt box, and Kora weaves in exactly the memories that matter for this question — the right parts, not your whole history, and you see what was added before you send.
And beyond the click, the memory simply travels with you: start a thought in one AI and carry on in another mid-sentence. Tell one AI something once, and every AI you use just knows it.
- Enhance with Kora — one click, and your prompt arrives fully briefed. Works in ChatGPT, Claude, Gemini and more.
- Continue anywhere — pick up in Claude what you started in ChatGPT, without re-explaining a word.
- Ask Kora — one question, answered from everything you’ve explored.
- Saves the re-explaining — and the tokens you’d have burned doing it.
How and why
to use it.
Not every question needs the same amount of firepower.
When a question is genuinely hard, Kemory can convene Counsel mode for you — several models answering in parallel, on the same question, at once. Disagreements get surfaced, not hidden, and what comes back is one synthesised answer, not four you have to sort through yourself.
When it’s a light question, Kora, your Chief AI Assistant, routes it to an open model instead. No frontier rates are used for what could be a simple grocery run.
More reasoning power exactly when a question warrants it — and none of the cost when it doesn’t.
Permissioned,
auditable AI memory.
Every memory is permissioned, every change is provable, and the whole thing can run inside your own walls. We do not train on your data. Delete anything, and prove it is gone.
Compliance claims will land here once they are certified — for now, this reflects what is built today.
Each entry is chained to the one before it — an altered record breaks the chain from that point on.
A new map for your
knowledge landscape.
Kemory is a permissioned memory layer that sits across every AI you use — not inside any single one of them. Tell one assistant something, and, subject to the permissions you set, the rest can draw on it too.
It doesn’t store a pile of chat logs. It runs a process that compresses your history down to contextualised summaries. The result behaves less like a search index and more like a model of you — what you decided, what mattered, what changed and why. Fully connected, auditable, and permissioned by you.
What are you
using it for?
One memory, three jobs, one promise — and four doors in.
Memory that survives the session, the repo and the machine.
Your coding assistant forgets the project between sessions, so you end up explaining your own codebase back to it.
- Pick up where the last session stopped. No re-reading your own work to remember what you were half-way through.
- Decisions stay decided. Why that library, why that pattern, what you already tried and threw away.
- Move machines mid-task. Begin on a laptop, carry on wherever it runs next.
- Every client, one memory. Claude Code, Cursor, Claude Desktop, Warp — one config entry each, the same memory behind all of them.
- The next agent knows what the last one did. Hand-offs stop being briefings.
For coding
Memory that survives the session, the repo and the machine.
Your coding assistant forgets the project between sessions, so you end up explaining your own codebase back to it.
- Pick up where the last session stopped. No re-reading your own work to remember what you were half-way through.
- Decisions stay decided. Why that library, why that pattern, what you already tried and threw away.
- Move machines mid-task. Begin on a laptop, carry on wherever it runs next.
- Every client, one memory. Claude Code, Cursor, Claude Desktop, Warp — one config entry each, the same memory behind all of them.
- The next agent knows what the last one did. Hand-offs stop being briefings.
For everyday work
Stop briefing your assistant every morning.
The hardest part of prompting was never the asking. It is the explaining.
- Start every conversation mid-conversation. The project, the constraints and the decisions are already there.
- Your standards, stated once. House style, British English, the way you want things laid out — not re-typed every session.
- Long projects stay coherent. Work that runs over weeks does not reset because a chat got too long.
- Ask your own memory a question. Kora answers from everything you have worked out, and shows where it came from.
- One click, fully briefed. Enhance with Kora weaves in the parts that matter for this question, and you see them before you send.
Across several AIs
Stop being the go-between.
You are the integration layer between six brilliant machines, and you are doing the job with a clipboard.
- Use the best model for each job. Draft in one, picture in another, code in a third. The thread survives the switch.
- Finish somewhere other than where you started. Mid-thought, without pasting a brief.
- Nothing to copy, paste or carry between them. The memory is already on both sides.
- Ask several at once. Counsel puts one question to several models, surfaces where they disagree, and returns one answer.
- Send less to say more. The distilled version travels, not the whole pile.
Your memoryshouldn’t scatter.
Every year brings better models and you will keep trying them. The default outcome is a spaghetti junction — your thinking spread across a dozen systems, none of them talking to each other, none of them yours.
- Try any AI without starting again. The next one opens already understanding the work, so switching costs you nothing.
- Judge the model, not its ignorance. A new AI looks worse than it is, because on turn one it knows nothing about you.
- One place, not a dozen. Your understanding sits above the models rather than inside each of them.
- See everything it holds, and delete any of it. Deleted means gone, and provable.
- If you leave, you take it with you. Export whenever you like. We do not train on it.
Who it’s for.
01
For anyone who uses AI
You do not need to be technical and there is nothing to set up. Add Kora for Chrome, carry on using the AIs you already use, and they stop forgetting.
Free to start — 2,000 memories, 2 connected AIs.
02
For agent-builders
The agents you build need memory. We built that once. You get to skip it. Multi-tenant and MCP-native, an identity and a rule per agent, versioned records.
Self-host with Docker or use hosted.
03
For AI companies
Your users arrive already known. One integration over MCP, the keys stay with the user, deploy where the data has to live.
Partner integrations available now.
04
For business
The same memory for a whole organisation — shareable namespaces, permissioning, audit, SSO and governance at scale.
Coming this autumn — taking design customers now.
Free to start,
hosted when you need it.
Free up to 2,000 memories. Pro at $19.99 a month. Enterprise adds governance, audit and support at scale.
Plans & pricing
| Feature | Free | Pro | Enterprise |
|---|---|---|---|
| Price | Free | $19.99 / mo | Custom |
| Memory | 2,000 / 2 AIs | Fair Usage Policy | Uncapped |
| Namespaces + Gatekeeper | Included | Included | Included |
| Hash-chained audit | Included | Included | Yes + export |
| Cross-AI capture | Included | Included | Included |
| Counsel & Kora | Included | Included | Included |
| Self-host | hosted + community edition | Not included | Yes / own-cloud / air-gap |
| Support | Community | Priority | Dedicated + SLA |
Free
- Price
- Free
- Memory
- 2,000 / 2 AIs
- Namespaces + Gatekeeper
- Included
- Hash-chained audit
- Included
- Cross-AI capture
- Included
- Counsel & Kora
- Included
- Self-host
- hosted + community edition
- Support
- Community
Pro
- Price
- $19.99 / mo
- Memory
- Fair Usage Policy
- Namespaces + Gatekeeper
- Included
- Hash-chained audit
- Included
- Cross-AI capture
- Included
- Counsel & Kora
- Included
- Self-host
- Not included
- Support
- Priority
Enterprise
- Price
- Custom
- Memory
- Uncapped
- Namespaces + Gatekeeper
- Included
- Hash-chained audit
- Yes + export
- Cross-AI capture
- Included
- Counsel & Kora
- Included
- Self-host
- Yes / own-cloud / air-gap
- Support
- Dedicated + SLA
Honest comparisons.
Even where we lose.
Memory is a crowded room. Here is where Kemory fits — including who to choose instead when they are the better fit.
Where Kemory sits
A map of notebooks, stores, and platform memory — so you can see the tradeoffs before the feature table.
Feature comparison
memory
memory
Kemory
- Works across every AI
- Yes
- Shared agent memory, permissioned
- Yes
- Builds a model of you
- Yes
- Ask many models / routing
- Yes
- You hold the keys (community edition only)
- Yes
- Integrated with SeKondBrain's Cognition Layer
- Yes(Coming Soon)
- Open-source
- Yes(community)
Nessie
- Works across every AI
- Yes
- Shared agent memory, permissioned
- —
- Builds a model of you
- Partial
- Ask many models / routing
- —
- You hold the keys (community edition only)
- Partial
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- No
Supermemory
- Works across every AI
- Yes
- Shared agent memory, permissioned
- —
- Builds a model of you
- Partial
- Ask many models / routing
- —
- You hold the keys (community edition only)
- Partial
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- Yes
Mem0
- Works across every AI
- —
- Shared agent memory, permissioned
- Flat
- Builds a model of you
- —
- Ask many models / routing
- —
- You hold the keys (community edition only)
- Partial
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- Yes
Honcho
- Works across every AI
- —
- Shared agent memory, permissioned
- Single-user
- Builds a model of you
- Yes
- Ask many models / routing
- —
- You hold the keys (community edition only)
- —
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- Yes
ChatGPT memory
- Works across every AI
- No
- Shared agent memory, permissioned
- —
- Builds a model of you
- Partial
- Ask many models / routing
- —
- You hold the keys (community edition only)
- —
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- No
Zep
- Works across every AI
- —
- Shared agent memory, permissioned
- Per-user
- Builds a model of you
- Partial
- Ask many models / routing
- —
- You hold the keys (community edition only)
- Partial
- Integrated with SeKondBrain's Cognition Layer
- —
- Open-source
- Core only
Where each one fits
Credit where it’s due. The real difference, either way.
Choosing one of these instead? Good — it means it fits you better right now. Try Kemory now for when cross-AI, governed memory is what you need.
Stop re-explaining yourself
to every AI.
Internally benchmarked at 87.8% on LongMemEval-500 end to end, using open-source models. Built for MCP — Claude Code, Cursor, Claude Desktop. Open models, your walls. Community edition and free plan coming soon.
Home
Home