Grace AI is an AI agent that runs in the cloud and is there for you around the clock. Reach it via the Grace AI app. Your secrets stay out of the LLM, your work is never training data, and technical tasks are held to real engineering standards.
See the options
Most AI agents hand your passwords and keys straight to the LLM. Grace AI keeps them in Vault, where the agent can use a secret without ever seeing it.
Passwords and keys go into Vault. The agent acts through them but never reads them into the LLM. Your instance is yours, not a slot in shared infrastructure where another customer's traffic sits beside your own.
Your data sits encrypted at rest and is only decrypted when the agent actually needs it. Backups run on a schedule, are encrypted separately, and we run restoration tests so a backup is something you can count on. The standard is simple: your work is still there tomorrow.
Context is managed on purpose and tools are chosen with intent. The agent is not built to burn tokens to look busy. Connect your own LLM keys and pay the provider directly if you prefer.
When the task is engineering, the agent works against explicit requirements: tests, constraints, acceptance criteria. The standard comes from people who ship production software.
We run a dedicated instance for you. Pick a plan by the monthly token allowance you need, or connect your own LLM keys and pay your provider directly. One person, one workspace, direct support.
From a low monthly price, pricing at launch
A dedicated instance for a team, built so people can work side by side without overwriting each other. Priced per instance, not per seat: add the people you need. Stronger hardware and higher support on the upper plans.
Per instance, pricing at launch
Buy the license, deploy on your own hardware or VPS, keep what you paid for. All updates within the major version are included. Available as a Personal or Business license.
Perpetual license, pricing at launch
Grace AI runs on the Grace backbone, the same engineering that powers Grace Mail and the rest of the suite: a real storage engine, runtime, built for speed and privacy. Infrastructure-grade tools with the accountability.
Managed or self-hosted, the runtime belongs to you. Configuration, retention, and access boundaries can be changed just for you.
Work from the Grace AI app on your computer (and soon from your phone). The agent runs in cloud, you can reach it from anywhere.
Grace AI is built and operated by Profit Visions s.r.o., a registered Czech company. GDPR by design. When managed hosting needs to sit in another region, that is a conversation, not a policy wall.
General work: research, writing, planning, code, operations. One AI agent across many contexts. What sets it apart is how it handles secrets, backups, tokens, and technical tasks.
Managed, Personal: one user, we run it, you pick a token allowance.
Managed, Business: a team on one instance, built for safe collaboration, priced per instance.
Self-hosted: you buy a license and run it on your own infrastructure.
Same product underneath. Different operational model.
Yes, on every option. Connect keys from LLM providers you already trust and pay them directly, or use a token allowance through Grace AI. You switch between them in settings whenever you like.
Vault is where Grace AI keeps your passwords, keys, and other secrets. The agent can act on your behalf using them without the secret ever reaching the LLM. You decide what goes in. Your files are separate: the LLM works with the files you give it, Grace AI just does not push everything to it automatically.
No. Your tasks, files, and messages are not training data for Grace AI. When you bring your own LLM keys, only that provider's own terms apply to those calls.
You take your configuration and data with you. On Self-hosted you already hold everything.