Never start from scratch with your AI again.
A memory that gets better with age, not worse.
Say it once in any AI. Every other one has it from your next conversation. Your context follows you between tools and stays yours on whatever model comes next.
Cancel anytime before day 7 and you won't be charged.
6,018 users · Customers in 70 countries · 65 started a plan in the last 30 days
6,018 users · Customers in 70 countries · 65 started a plan in the last 30 days
Works inside the AI tools you already use
.png&w=3840&q=75)
.png&w=3840&q=75)
You keep re-introducing yourself.
Every new chat, every new tool, every model launch: back to zero. People report spending 3 to 4 hours a week just re-explaining context their AI should already have.
Brief your AI on the project
Repeat it in the next chat, the next tool, the next model
Lose hours every week to context you've already given
What you need isn't another AI. It's one memory they all share.
Tell Claude once. ChatGPT already knows.
You
Assistant
Saved to memory · rebrand: Fern & Field
memory
You
With MemoryPlugin
Recalled from memory · rebrand: Fern & Field
3-4 hrs / week
what people report spending re-explaining context to AI, again and again
< 1 of those hours
what MemoryPlugin costs per month
Built for how you actually use AI.
Pick your lane: coding, writing, decisions, client work. Same AI, except it finally knows yours: your stack, your voice, your clients.
Your AI knows your stack, conventions, and the codebase you've been building in, without you re-pasting it every chat.
Without memory
“Set up a Next.js project. Use TypeScript strict, Tailwind v4, the folder layout from my last project, and our team's commit style…”
With MemoryPlugin
“Scaffolding Next.js with TypeScript strict and Tailwind v4, using your standard /app /lib /components layout. I'll follow the conventional-commit style you used in the last repo.”
One price, every AI tool. 14-day refund, no questions.
One memory, two layers.
The notebook
What you choose to keep. Curated memories, organized in buckets, with your AI in every conversation.
The archive
Everything you have said. Years of conversations, searched the moment they matter.
Both live outside every app, and both follow you everywhere.
Import years of AI conversations.
Your AI is only as good as the context it has. You've already had thousands of conversations: pricing debates, project plans, personal reflections. MemoryPlugin imports them from 6 sources in a few clicks. No scripts, and it keeps syncing on its own.
Scanned
conversations, 5 most relevant selected automatically
Injected
tokens of rich personal context, details you'd forget to mention yourself
Retrieved in
vs. ~45 min to gather manually, if you even could
Retrieved from 5 conversations
Pricing experiment results: 4.2% conversion at $29
Chose paid-first: free users created 70% of support tickets
Customer feedback: “I’d pay $50/yr for this”
Runway calculation: need revenue within 6 months
Competitor analysis: everyone does freemium poorly
You
Without memory
· GenericWith MemoryPlugin
Stop repeating yourself to AI.
Give your AI the context it needs to actually be useful.
Code with AI? Get the Mac app to bring your Claude Code, Codex, and Cursor sessions along.
Get StartedA memory that organizes itself.
You barely feed it. It keeps itself clean. It answers back in every AI you use. Context engineering, done for you.
Capture
Saved without trying.
As you chat, the AI saves what is worth keeping. Highlight anything to capture it in one click. Buckets keep work, personal, and projects apart; share one with someone you work with.
Keep clean
It curates itself. You approve.
Duplicates get merged, stale facts get flagged, and you approve every fix. Smart Memory then injects only the context each conversation needs. It is why answers get sharper the longer you use it.
Answer back
Ask your memory anything.
Your memories, uploaded files, and years of chat history live in one place. Ask in plain language and get a real answer with its sources. Long chats become summaries; months become insights.
“Doesn't ChatGPT already have memory?”
It does. So does Claude. So does Gemini. And none of them share: each remembers a different slice of you, locked inside its own app. What they do remember looks like this:
Junk
the same fact, saved three times
Full
capped right when it knew you
Stale
you moved to Lisbon months ago
Rewritten
your details, averaged into a vibe
MemoryPlugin is the opposite: one memory outside every silo, curated instead of hoarded. Add to it in any tool; every other tool has it from your next conversation.
Want the long version? The full comparison with built-in memory.
Built-in memory forgets. Yours compounds.
Loved by people who use AI every single day
Press, Reddit threads, and verified customers: here's what people actually say after using it.
“Like an additional super-brain for AI assistants.”
I've been using MemoryPlugin daily for the past year, and it has dramatically increased my productivity when working with large language models. Less AI blah-blah, more valuable text, no matter which LLM it is, the difference is drastic. It is like an additional super-brain for AI assistants.
The features of MemoryPlugin are constantly evolving and consistently outperform the built‑in long‑term AI memory solutions offered by the major providers.
“MemoryPlugin makes AIs a flexible field instead of a marriage. I've become dependent on it — world-changing product.”
“Elegantly solves AI amnesia while maintaining user control and data protection.”
“DeepSeek + OpenRouter + TypingMind + MemoryPlugin… and I have my 4o back. Really, THANKS.”
“Pays for itself if it saves you even 30 minutes a month.”
“Omg, this plugin is saving me hours of work and helping me to track habits and moods. Thanks again for this amazing tool. It really is incredible.”
“I'm still using MemoryPlugin, which is superior in every way.”
“So far I'm thoroughly happy with MemoryPlugin. It has greatly helped me with my work by expanding the long-term memory functions, so I don't have to fully repeat my work in every new chat. That's such a time saver.”
“If you're on the fence about MemoryPlugin, don't hesitate — this extension is an absolute game-changer. The functionality is UNREAL. It's not just useful, it's essential. Worth every penny, and then some.”
“Thanks to this plugin, I'm working much more comfortably. It really helps a lot, 10/10.”
The only memory you need.
The whole memory system, with a few power features tagged Pro. Tap any feature for the short version.
The notebook
The archive
Ask
Everywhere
Coming soon
Trust
Your memory is yours.
You’re trusting it with real context about your life and work. We built for that responsibility.
Encrypted in transit and at rest
Never sold, never used to train models
You choose what it remembers
Yours to export and delete, anytime
We even answer the uncomfortable one, honestly: “Can you read my chats?”
Pricing
One subscription that makes every AI better
Core is the complete everyday memory. Pro adds your entire history, and the intelligence on top.
Pro Lifetime · next 5 only · ends Sep 21
The next five Pro Lifetime seats are $100 off.
$300 once instead of $400. Less than two years of Pro, and nothing to renew, ever. The code VKDNCV07GN is applied for you at checkout.
Core
The everyday memory system.
One memory for every AI you use
- Unlimited memories, across 21+ AI tools
- Buckets, suggestions, and Smart Memory included
- File Q&A, image memories, version history
- Chat history: your last 500 conversations
7-day free trial · 30-second setup · 14-day refund
Pro
Plus your entire history, understood.
Access everything you've ever discussed with AI
- Everything in Core, plus:
- Your entire chat history, unlimited
- Knowledge graph of everything you've discussed (early beta)
- Conversation summaries, monthly insights, usage analytics
- Shared buckets for your team
7-day free trial · 30-second setup · 14-day refund
6,018
Users
70
Countries
4.5★
From 100+ reviews
21+
AI tools
65 people started a plan in the last 30 days, the most recent 23 hours ago.
Private by design · Your data stays yours · Export everything, anytime
Six differences. Everything else comes with every plan.
The full feature list is one scroll up.
Chat history import
CoreLast 500 conversations
ProUnlimited
Knowledge graph (early beta)
Core—
ProIncluded
Summaries & monthly insights
Core—
ProIncluded
Usage analytics
CorePreview
ProFull access
Shared buckets
Core—
ProInvite your team
New features
CoreStandard rollout
ProPriority access
Price
Core$180$89 / year
Pro$300$180 / year
Chat History Import: Core includes your last 500 conversations; Pro is unlimited. Imports work from ChatGPT, Claude, Gemini, TypingMind, Grok, and DeepSeek on every plan.
Works everywhere you AI
Switch models mid-thought; your memory comes with you. The widest platform coverage anywhere, from ChatGPT and Claude to whatever drops next.
One extension.
.png&w=64&q=75){kind=small}
.png&w=64&q=75){kind=small}
Not just another MCP server.
Most memory tools connect one way and are built for developers: a Docker image here, a Markdown vault there. MemoryPlugin is built for people who just want their AI to know them, and it connects every way that matters, right where you already work.
Your coding agents join your one memory. It syncs your Claude Code, Codex, and Cursor conversations into chat history, so any AI can recall what you decided in the terminal.
One open-source install teaches your agents to recall your context before assuming and save what matters before finishing. Memory becomes a habit, not a feature.
What everyone asks before signing up.
For setup guides and deep-dives, browse our knowledge base ↗.
MemoryPlugin is a long-term memory layer for the AI tools you already use: ChatGPT, Claude, Gemini, Grok, DeepSeek, Perplexity, Mistral, TypingMind, Claude Code, Cursor, Windsurf, and more. It captures memories you (or the AI) want to keep, imports your existing chat history, and lets you upload reference files. It then surfaces all of that as context in future conversations, on every tool you use.
See how it works for the full picture.
It does, and so do Claude and Gemini. But each one is locked inside its own app, so every AI remembers a different slice of you. Built-in memory also fills up, can get wiped, and decides for itself what is worth keeping.
MemoryPlugin keeps one memory you control, shared across every tool it supports, organized into buckets, with your full chat history searchable from any AI. Add to it in one tool and every other tool has it from your next conversation. Built-in memory is a start. MemoryPlugin is the memory that follows you.
Yes. The 2026 memory overhauls replaced the list of saved memories with a rolling summary that updates on its own, and plenty of people miss the old way: exact facts, saved on purpose, that stay put. That is how MemoryPlugin has always worked.
Export your ChatGPT memories with our free export tool, import them in a few clicks, and they become discrete memories you edit, organize into buckets, and use in whichever AI you open. Nothing gets rewritten behind your back. See the full comparison with built-in memory.
Those are good tools aimed at a different user. OpenMemory runs as a local Docker service, Basic Memory is a Markdown knowledge base you organize yourself, and Supermemory has shifted toward developer and enterprise infrastructure. MemoryPlugin is the consumer product: install the extension or connect the MCP server, and one curated memory works across your tools with nothing to host or maintain.
Detailed comparisons: vs Supermemory, vs OpenMemory, vs Basic Memory, vs built-in memory.
Pick the path that matches your setup; you can mix freely:
- Browser extension: works across ChatGPT, Claude, Gemini, Grok, DeepSeek, Perplexity, Mistral, Poe, LibreChat, Qwen, and 6+ more. Setup guide.
- MCP server: for Claude Code, Cursor, Windsurf, Claude Desktop, and 100+ other MCP clients. Setup guide.
- Custom GPT: fastest path for ChatGPT, also works in the iOS and Android apps. Setup guide.
- TypingMind plugin: native integration. Setup guide.
- API: build your own integration. API docs.
See the full list of supported platforms.
Yes. MemoryPlugin Sync is a free macOS app that brings your coding agents into your one memory: it syncs your Claude Code, Codex, and Cursor conversations into your chat history so any AI can recall them. Download it here. Claude Code, Cursor, and other coding tools also connect live through our MCP server.
Yes. The Chrome extension runs on Android via Quetta Browser, Safari on iOS has a dedicated extension (linked from your dashboard), and the TypingMind plugin works in their iOS and Android apps. ChatGPT's mobile and desktop apps work via our Custom GPT.
Yes. Memory entries store just the memory text — not your full chat. Chat History import is opt-in: when you choose to import, the full conversations are stored so your AI can search them later, but nothing is uploaded without your explicit action.
Everything is encrypted in-transit and at-rest, kept private and separate from other users, never sold, and never used to train models. You can delete any memory or conversation at any time from the dashboard.
Want the full, no-spin version, including “can you read my chats?” answered honestly? It's on our trust page. More detail in the Chat History overview and our privacy policy.
You stay in control. Use the Exclude option in your dashboard to remove specific conversations from search and context — when you exclude a chat, its search index is deleted and it won't be re-imported on future syncs. The stored conversation itself stays until you delete it.
For live conversations, you can also disable injection per-chat in the browser extension, or toggle the integration off in TypingMind.
Details in using chat history.
Anytime. Edit or delete individual memories from the dashboard, run bulk operations, or delete a whole bucket. Deletes are permanent.
Yes. Direct file export is supported for ChatGPT, Claude, TypingMind, and Grok. Gemini and DeepSeek use online sync via the browser extension, which keeps new conversations syncing automatically after your initial import. Once processed, your AI can search across years of past chats.
See the import guide.
Yes. Every subscription starts with a 7-day free trial. Add your card to start, then cancel anytime before the trial ends and you won't be charged. After that you're billed for the plan you picked, still backed by our 14-day refund guarantee. Lifetime plans are a one-time purchase, so they don't include a trial.
Yes. If MemoryPlugin isn't a fit, reply to your purchase email within 14 days for a full refund. No forms, no questions. Have another question? Email us.
MemoryPlugin
One memory. Every AI.
Set it up once. Every AI you use remembers you, and the memory is yours to keep, wherever you work next.
Start Your Free 7-Day Trial7 days free. Cancel anytime before you're charged. 14-day refund after.
