Membase Chrome Extension
Save conversations and web pages as you browse, then bring relevant context into ChatGPT, Claude, Gemini, Grok, DeepSeek and Kimi.
One memory.
Every AI. Only yours.
Give your AI a lasting memory of what matters to you. Bring together what you know, learn and create. Decide what to keep and who can access it.
Connect the AI you use.
ChatGPT
Claude
Gemini
Grok
Kimi
DeepSeek
FOR DEVELOPERS
Build AI with memory that lasts.
Let your apps and agents build on accumulated knowledge, user preferences and past decisions. Connect through MCP, REST, Python or TypeScript, with control over what each can access.
$ npm i @unibaseio/membase1# pip install unibaseio-membase2# export MEMBASE_API_KEY="mbk_..." (Connect > Developer keys in the app)34from membase import Membase56client = Membase()78# Remember something9client.add(10 "Call notes with Acme: they want SSO before the pilot.",11 container="mv-...",12)1314# Search returns relevant passages. Use client.ask() for an answer.15hits = client.search("what does Acme need before the pilot", limit=5)16print(hits["results"][0]["content"])One place to build and manage your memory.
Bring in what matters, keep it up to date, and choose where to use it.
Membase
Memory Sources
Membase Chrome ExtensionAI conversations and web pages
Connected AppsThe apps you work in
Upload FilesPDFs, notes and images — or a zipped export of your notes
Your Own AgentsThrough MCP or the API
Put your memory to work.
MAKE AI PERSONAL
Give AI a memory of what matters to you.
Save your preferences, goals and working style so your AI can draw on what matters to you when you write, plan and work.
FIND WHAT MATTERS
Put accumulated knowledge to use.
Bring research, notes and past decisions together. Ask questions across your saved knowledge and use what you find in your next idea or task.
BUILD WITH AGENTS
Give agents knowledge they can build on.
Let authorized agents save findings and retrieve relevant knowledge as work progresses. Share selected memories across agents and keep access under your control.
Use Membase where you work.
Capture and reuse context in your browser today. Desktop access is coming soon.
Add to Chrome
Membase Desktop
Coming soonPlanned desktop features include a local copy of your memory, offline search and access for agents running on your machine.
YOUR MEMORY
Your memory. Your control.
Decide what stays in your memory, who can access it, and when to take it with you.
Decide who can use your memory.
Choose which memories to make available to each connected AI, and revoke access whenever you need to.
Review, correct and remove.
Keep your saved knowledge useful as your work changes. Correct mistakes and remove information you no longer want to keep.
Take your memory with you.
Export your memory and sources in documented formats, so you can keep a copy outside Membase.
ARCHITECTURE
How Membase handles your data.
Your account’s memory runs in an isolated environment. The hosted service stores and processes the sources you provide to build and retrieve that memory.
HOW YOU CONNECT
Web App
BUILD AND MANAGE
Chrome Extension
SAVE AND REUSE
API & MCP
CONNECT YOUR AGENTS
YOUR ACCOUNT'S BOUNDARY
YOUR MEMBASE ACCOUNT
Your sources
The files, pages and connected content you provide
Your isolated memory environment
Memory processing and retrieval scoped to your account
The hosted service is not end-to-end encrypted. For the Chrome extension’s data handling, read the extension privacy policy.
BENCHMARKS
Memory performance, measured.
Answer accuracy and context usage for the unibaseio-membase-core memory engine.
Answer accuracy (%)
Questions answered correctly on each benchmark. Higher is better.
LoCoMo
93.1%
1,540 questions
LongMemEval_S
92.6%
500 questions
DMR
92.2%
500 questions
Context tokens per question
Mean retrieved context compared with the full conversation history.
LoCoMo
~67% fewer tokens- Full history
- 19,987
- Membase context
- 6,562
LongMemEval_S
~91% fewer tokens- Full history
- 102,795
- Membase context
- 8,970
Each token chart is scaled to its own full history. Reductions are rounded.
These results measure unibaseio-membase-core, not the hosted Membase service. Reader models and evaluation methods vary by benchmark; see the report for details.
FAQ
- Membase is the memory layer for AI agents and the people who use them. It builds one long-term memory from your conversations, files and connected apps, and every AI tool you connect can read it, including ChatGPT, Claude, Cursor and your own agents. You choose what goes in and which tools can access it. Membase is built by Unibase.
- No. Membase (membase.ai) is the AI memory layer built by Unibase. It is not affiliated with membase.so, and it is unrelated to the Membase key-value database that later became part of Couchbase.
- The Membase Chrome extension works with ChatGPT, Claude, Gemini, Grok, DeepSeek and Kimi. Coding tools such as Claude Code, Cursor and Codex connect through the Membase MCP server. Custom agents can use MCP, the REST API, the CLI, or the Python and TypeScript SDKs.
- Install the SDK with pip install unibaseio-membase or npm install @unibaseio/membase, set your API key, then add and search memories in a few lines of code. Any MCP client can also connect to the Membase MCP server directly. The quickstart is at docs.membase.ai.
- Membase answers 93.1% of LoCoMo, 92.6% of LongMemEval_S and 92.2% of DMR questions correctly. It sends the model about 67% fewer tokens than the full conversation history on LoCoMo, and about 91% fewer on LongMemEval_S. The full method and results are on the Membase benchmarks page.
- Both Membase and Mem0 give AI agents long-term memory through an API. Membase is built around one memory that you can use from consumer AI apps (through the Chrome extension) and from developer tools (through MCP and the SDKs). It runs hosted or locally, and its benchmark method and harness are published.
- Yes. Membase runs hosted, or on your own machine with pip install 'unibaseio-membase[local]' (Python 3.12 or 3.13, with your own OpenAI API key). The same API and MCP tools work in both modes.
- The Membase SDKs, CLI and MCP server are public on GitHub at unibaseio/membase-ai. The core memory engine is proprietary and ships as a compiled package under the Unibase license.
- Yes. You choose what gets added: synced conversations, saved items, uploaded files, connected apps, or knowledge added by authorized agents. You can keep separate memories for different projects, review and correct what was saved, delete or export it, and revoke any tool’s access at any time.
- The hosted service stores and processes the sources you provide in order to build and retrieve your memory. Each account’s memory runs in an isolated environment. The hosted service is not end-to-end encrypted, so use local mode if your data must stay on your own machine. The Chrome extension has its own privacy policy.
Comparing options? Best AI Agent Memory Tools in 2026: 9 Options Compared
Build a memory that grows
with you.
Bring what matters into Membase. Keep what you learn, build on what you know, and give your AI the memory to help you move forward.