Ambient recall on your Mac.
Or conversations you can carry forward.
Minimi, from Shram Intelligence, is a Mac menu-bar app that remembers what crosses your screen, keeps it locally and nudges you about open loops. Kemory captures your AI conversations and serves one permissioned, audited memory to every AI, agent and colleague you choose.
Last verified 7 November 2026 against primary sources
Two good answers.
Different jobs.
- You want memory of everything you do on your Mac, not only AI chats, with little to set up.
- You want proactive nudges about commitments and open loops.
- Keeping stored memory on your own device is your top privacy requirement.
- You want whole AI conversations captured across ChatGPT, Claude, Gemini, Perplexity, Manus and Grok, on any computer that runs Chrome.
- You need to share memory with colleagues or other organisations under deny-by-default permissions, consent and a hash-chained audit log.
- You want to continue a conversation in another AI, ask across everything with sources, or get one answer from several models.
See each one
doing its job.
Kemory screenshot taken 7 November 2026. Minimi is a trademark of its owner, named here to identify the product.
At a glance
The same rows on every comparison, so you can read one page against another.
*89.6% end-to-end answer accuracy on LongMemEval-S, open-weight Llama-3.3-70B reader, six runs. See the benchmarks or the full method on docs.sekondbrain.ai. A dash means not documented, not proven absent.
The same job,
done two ways.
Minimi captures whatever crosses your Mac's screen, including mail, documents and chat apps that a browser extension cannot reach. It starts working with little to configure, and it pings you about open loops rather than waiting to be asked.
The numbers,
and how each was run.
These rows use different datasets, splits, readers and retrieval budgets, so they cannot be compared directly. We publish the method, every run including the lower ones, and use an open-weight reader rather than a frontier model.
| Product | Benchmark | Score | What it measures | Setup | Reader model | Runs |
|---|---|---|---|---|---|---|
| Kemory | LongMemEval‑S | 89.6% | End-to-end answer accuracy | 500 questions, oracle split | Llama-3.3-70B (open-weight) | Six, with variance |
| Minimi | BEAM (ICLR 2026) | “50% more accurate” than 36% SOTA; one page shows 54% | Vendor-reported | Not stated | Not stated | Not stated |
What changed at Minimi
Products move quickly. These are the dated changes this comparison has been checked against.
- 5 June 2026Launched on Product Hunt, spun out of the memory layer behind Shram.
- Since August 2026Repositioned as a proactive “AI cat” that pings you; MCP support extended from Claude to ChatGPT and Gemini.
- Since August 2026“Company brain on Claude” team page added.
Specialists do one job well.
Kemory holds the thread across them all.
Every year brings better models and you will keep trying them. Kemory sits above the models rather than inside any one of them, so what you have worked out goes with you.
Tell one AI something once, and every AI you use just knows it. Start a thought in one and carry on in another, mid-sentence.
Every memory is permissioned and every change is provable. We do not train on your data. Delete anything, and prove it is gone.
Compaction runs on an open-weight model, so the frontier model only ever sees the compacted version — not the pile.
Already using Minimi? You do not have to replace it to try Kemory. Kemory connects to the AIs you use over MCP and Kora for Chrome.
Choosing Minimi instead?
Good — it fits you better right now.
Come back to Kemory when cross-AI, governed memory is what you need.
Five questions,
Kemory vs Minimi.
- What is the difference between Kemory and Minimi?
- Minimi is a Mac app that remembers what crosses your screen and plugs that memory into your AI over MCP. Kemory captures whole AI conversations across six AIs and serves one permissioned, audited memory to your AIs, agents and colleagues.
- Is Minimi entirely on-device?
- Memory is stored locally on your Mac, but Minimi says it generates embeddings through Google's Gemini API, so some processing leaves the device. Kemory holds data in India for Indian users and in Switzerland otherwise, with enterprise self-host available.
- How much does Minimi cost compared with Kemory?
- We found no prices on Minimi's site as of 7 November 2026. Kemory is free for 2,000 memories and two AIs, with Pro at $19.99 a month, Team at $30 a seat and Business at $55 a seat.
- Can Minimi and Kemory work together?
- Both connect to AIs over MCP, so you could add both to Claude. They capture different things: Minimi captures on-screen activity, and Kemory captures whole AI conversations with sharing and audit.
- How do Minimi's benchmark claims compare with Kemory's?
- Not directly. Minimi reports a BEAM result as “50% more accurate than previous SOTA”, without stating the metric or reader. Kemory publishes 89.6% end-to-end answer accuracy on the LongMemEval-S oracle split with an open-weight Llama-3.3-70B reader, across six runs.
More comparisons
Credit where it’s due. The real difference, either way.