Home / Products
One agent. Three places to run it.
wAIve is the agent we build and use every day. Run it in the browser on our lab, install it on your own computer, or put it in your terminal to write code. Point the last two at any model.
wAIve.online
Agent workspace in the browser. Models self-hosted on our lab. Start free.
Learn more →wAIve Desktop
The whole agent on your computer, with local models or almost any provider.
Learn more →wAIve CLI Early access
The coding agent in your terminal, on your own models or any OpenAI-compatible server.
Learn more →Which one fits?
| wAIve.online | wAIve Desktop | wAIve CLI | |
|---|---|---|---|
| Where it runs | Browser, hosted on our lab | Your computer (desktop app) | Your terminal (Linux; Windows via WSL) |
| Your files live | Private cloud workspace | A folder on your disk | Your git repo |
| Models | Flash / Ocean / Tidal, self-hosted on our lab | The mTec Engine or any model you connect | Any OpenAI-compatible server; needs tool calling and 32K+ context |
| Modes | Chat Mode + Agent Mode | Agent Mode, plus long-task, permission and background features | Coding agent with an autonomy dial: plan, manual, normal, auto |
| Sign-in | Account; free plan and paid plans | No account needed | No account needed |
| Runs in the background | In the cloud | In the tray | Goal loops until your tests pass |
| Best for | Getting started, any device | Power users, local models, private files | Developers and dev teams |
wAIve.online and wAIve Desktop run on the mTec Engine when you use our models. wAIve CLI is in early access. Want any of them on hardware you own? See Hardware.
Start with a free exploration call.
Tell us what you're trying to do and what's in the way. If AI fits, we'll show you where. If it doesn't, we'll say that too.