pleeply Chat
One interface for different AI models. Switch models in the same conversation and see where every answer was computed.
One AI ecosystem on your computer. Run models locally, connect to the cloud when you choose.
Preview of pleeply Desktop, which is in development. What you type here isn’t sent or stored.
Nothing is released yet. “First build” means planned for the first early-access build. “Later” means on the roadmap after that.
One interface for different AI models. Switch models in the same conversation and see where every answer was computed.
Discover, download and manage local and cloud models. Each one is checked against your hardware before you download it.
Manage hardware, performance, privacy and model settings in one place.
Personal AI agents for research, productivity and automation, with a check-in before any step that matters.
A personal context system that helps AI understand how you work, stored on your computer.
Local when it can. Cloud when you say so.
Three modes, one switch. pleeply is designed to show where every request runs, and to ask before anything leaves your computer.
Saturn, Nova, Atlas and Orbit are names for model setups we plan to offer inside pleeply. None of them is released yet, and they are not foundation models we trained from scratch.
The all-rounder for conversation, writing and reasoning, sized to what your computer can run.
Quick answers with a small memory footprint, made for instant replies and modest laptops.
Reads long documents and notes, pulls out what matters and points to where it came from.
Plans multi-step tasks and works with your tools, checking in before actions that matter.
Install the desktop app when your early-access invite arrives.
It reads your chip, memory and graphics to see which models will run well.
Pick from the models that fit, or connect a cloud model when you need more power.
Your data should belong to you. Private by default where that’s possible, honest where it isn’t.
Compatible models run on your computer. In Local mode, prompts and files stay on your device.
Cloud models are an add-on, never a requirement. Connect a provider only if you want more power.
Every answer shows where it ran: on your computer or with a named cloud provider.
pleeply is designed to ask before the first request goes to any provider, and cloud can be switched off entirely.
Not yet. pleeply Desktop is in development. Join early access and we’ll email you when invites start. We won’t promise a date we can’t keep.
No. What runs locally depends on your memory and graphics. Small models (roughly 1–8B parameters) run on many recent laptops with 16 GB of memory; larger ones need more memory or a strong GPU. pleeply is designed to check your hardware and recommend models that fit.
In Local mode, no: prompts and files are processed on your device. When you choose a cloud model, your request goes to that provider and their privacy policy applies. pleeply is designed to ask before that happens.
Planned names for model setups inside pleeply, each aimed at a different kind of work. None of them is released yet, and they are not foundation models we trained from scratch.
We haven’t set pricing for pleeply Desktop. Joining early access is free and doesn’t commit you to anything.
A small device that would run pleeply at home. It is in design, with planned prices from about $299 to $549. There’s no waitlist for it yet: pleeply Desktop comes first.
The desktop app comes first. pleeply Pocket comes later, and there’s no waitlist for it yet.
Be among the first to try the app. We’ll invite people in waves, and your answers help us test on the computers people actually have.
We’ll email you when invites start. Know someone who’d want it? Share your link.
A small device that runs pleeply at home, no computer needed. There is no working unit yet, and prices are planned.

Planned prices. No waitlist yet: we’re building pleeply Desktop first.