It remembers everything, so you don't have to. A private memory of your documents, mail and notes. It runs on your machine, powered by your own LLM.
Free, forever, on your own hardware.
One quiet promise: the data stays yours, the model stays yours, and the context is the one you choose.
Your data never leaves your machine. The app talks only to a local sidecar and the AI runtime you point it at.
Bring any OpenAI-compatible runtime: LM Studio, Ollama, vLLM or llama.cpp. No vendor lock-in.
Index mail, calendar and folders so Hygur can answer with your own context, not the public web.
Hygur does the remembering, so you move faster. It reads your own world, ranks what matters, and brings it to you before you go looking.
Each morning, what moved and what needs you today, ranked so there's nothing to triage.
Decisions to make, deadlines and follow-ups rise to the top before they slip.
Hygur keeps what you decided and why, and flags it when a later source disagrees.
Recurring bills, renewals and obligations surface before they fall due.
An assistant worth trusting is one you can check. Hygur answers from your own world, shows its sources, and tells you when they disagree.
Every reply cites your own mail, notes and documents. Never the public web, never invented. Dates and amounts are computed, not guessed.
When two of your own sources disagree on the same fact, Hygur surfaces it, both values quoted and dated, so you decide what still holds.
It runs on your machine, or a private EU instance that's yours alone. No training on your data, ever.
Run it free on your own machine, or let us host it for you on EU servers. Connect your world with a growing catalogue of connectors. Your data stays yours the whole way.
Desktop and mobile
The local twin in your pocket and on your desk. Ask, search and capture across Mac, Windows and your phone. It runs on your machine and talks only to the AI runtime you point it at, local or any OpenAI-compatible endpoint.
Fully managed · 29.90 €/mo launch, then 35 €/mo
A managed Hygur instance, one per account, running exclusively on EU servers. GPU inference runs at an EU provider that never trains on your data. We run and update it; you keep control of your data, and can export or delete it anytime.
Connectors catalogue
A growing catalogue of connectors that bring more of your world in. Mail, files and calendars today; more tools to come for a richer memory.
Shared memory for teams & projects
Everyone keeps their private twin. A project gets its own shared memory, a mesh of connected minds where the team's mail, notes and decisions live together, with provenance and a daily brief.
Download the app, point it at a local LLM you choose, and connect your mail, files and calendar. Everything is indexed and answered right here, on your own hardware. No account, no cloud.
Runs on your Mac, Windows or phone. Open it and it's ready; there's nothing to set up in the cloud.
Connects to a local LLM you choose: LM Studio, Ollama, vLLM, llama.cpp, or any OpenAI-compatible endpoint.
Your mail, files and notes are indexed on your machine and answered by your model. Nothing leaves your side.
Free on your own machine today. When you want it hosted, a managed Hygur Cloud instance gives you the same Hygur on EU servers, built from the same sources you connect.
Hygur Cloud runs exclusively on EU servers, with GPU inference at an EU provider that never trains on your data.
Get Hygur