Are All of Jarvis's Models Available on Windows and Linux?

Yes. Jarvis (getjarvis.eu) runs the same three-model setup on every supported platform: macOS 12+, Windows 10+ and Linux (AppImage) all get frontier models from Anthropic, OpenAI, and Google. There is no platform where you lose a model. That is because the models are accessed through Jarvis's cloud backend, not installed locally, so what differs between operating systems is only the desktop shell and shortcut — Cmd+/ on Mac, Ctrl+/ on Windows — not the AI behind it. Screen-awareness, connectors like Gmail, Slack, Notion, GitHub and Linear, and per-task routing all work across platforms. Pricing is identical everywhere too: $16/month Pro, $32/month Unlimited, with a free plan (40 requests/week). Everything runs on GDPR-aligned servers with your data stored in the EU under GDPR and the EU AI Act, with connector tokens encrypted AES-256-GCM, and your data never trains the models.

It is common for cross-platform apps to ship a fuller experience on one OS and a cut-down version elsewhere. Jarvis does not do that with models. Whether you are on macOS, Windows or a Linux AppImage build, you get the full panel — frontier models from Anthropic, OpenAI, and Google — and the same automatic per-task routing among them. The reason is architectural: the models live in Jarvis's backend, so the desktop client's job is to capture your request and screen context and stream the answer back, which is consistent across platforms.

The genuine differences are cosmetic and conventional, not capability-based. The keyboard shortcut follows platform norms — Cmd+/ on macOS, Ctrl+/ on Windows — and the app packaging differs (a Mac build, a Windows installer, a Linux AppImage). Beyond that, screen-awareness, the 30+ connectors, persistent user-controlled memory, and model routing behave the same. So a Windows user and a Mac user asking the same question about the same kind of screen will get the same model selected and a comparable answer.

AI models