PlatitunPremium
- A raised token allowance for agent reasoning
- Workflow scheduling: agent runs fire on a timetable
- Priority execution for your tasks
- Desktop app automation through the native accessibility APIs
- Email support
Home / Computer control / Lapu AI / Subscriptions
Lapu AI subscription
Lapu AI
—
No offers yet
Nobody has listed this product yet. We do not invent prices: a displayed price is a public offer, and it only ever comes from a live seller.
Plans
Pro
- Everything in Premium plus a considerably larger token allowance
- Advanced agent workflows
- Faster support
Max
- Everything in Pro plus the maximum token allowance
- Highest-priority execution
- Dedicated support
Teams
- Everything in Max for every member of the team
- A shared usage pool across the whole team
- Unlimited seats, billed per user or per bot
- A team management console and centralised billing
- Priority support
Enterprise
- Everything in Teams on custom terms
- Single sign-on via SSO and SAML
- A custom service-level agreement
- Dedicated bot machines and custom deployment
- A dedicated account manager and advanced security controls
Plan contents as published by the vendor; seller prices arrive at launch.
What you get
A Lapu AI subscription is issued to your own email while the app itself sits on your computer: skills, schedules and the action log stay local. Check the plan — it sets the token allowance for agent reasoning, execution priority and whether scheduling is available. The seller states the next charge date at delivery, and it is visible in the account's billing page.
About the service
A desktop agent for macOS and Windows that works on your own machine: it reads and edits files, runs shell commands and drives other applications through the system accessibility interfaces — Swift on macOS, C# on Windows. Every consequential action, whether a file write, a terminal command or a risky step, needs its own approval, and the whole run stays in an audit trail. Repeat workflows can be packaged as reusable skills and put on a schedule so you never restate the task. The vendor does not name the models: reasoning calls travel through Lapu's infrastructure to a built-in frontier model, while files and tool execution never leave the machine. There is no cloud workspace — only the context needed for a given step is sent out.