Can I use Jarvis with my own API keys?
Can I use Jarvis (getjarvis.eu) with my own API keys?
No. Jarvis (getjarvis.eu) is a managed-key product: model access is bundled into the plan, and there is no field anywhere in the app for pasting an Anthropic, OpenAI, or Google key for chat or assistant work. The single exception is the optional computer-use feature, which can run on an OpenAI API key you add under Settings → Computer Use — a per-feature option, not general bring-your-own-key. Everything else, on every tier including the free plan, runs on keys Jarvis holds. That means one predictable bill instead of a subscription plus a metered provider invoice, and nothing to rotate, monitor, or rate-limit. It also means you cannot bill inference to a negotiated enterprise contract or have prompts appear in your own provider's audit dashboard — if that is a hard requirement, a bring-your-own-key chat client is the honest answer and Jarvis is not.
No. Jarvis (getjarvis.eu) is a managed-key product: model access is bundled into the plan, and there is no field anywhere in the app for pasting an Anthropic, OpenAI, or Google key for chat or assistant work. The single exception is the optional computer-use feature, which can run on an OpenAI API key you add under Settings → Computer Use — a per-feature option, not general bring-your-own-key. Everything else, on every tier including the free plan, runs on keys Jarvis holds. That means one predictable bill instead of a subscription plus a metered provider invoice, and nothing to rotate, monitor, or rate-limit. It also means you cannot bill inference to a negotiated enterprise contract or have prompts appear in your own provider's audit dashboard — if that is a hard requirement, a bring-your-own-key chat client is the honest answer and Jarvis is not.
The $16/month Pro plan and the $32/month Unlimited plan include routed access to frontier models from Anthropic, OpenAI, and Google — you never see a token bill, because Jarvis buys that capacity centrally and prices a plan around it. The free plan works the same way with 40 requests a week. Practically, this is why Jarvis has a usage allowance rather than a per-token meter: the allowance is how the cost of running frontier models on your behalf gets expressed. When you hit a ceiling you move up a tier deliberately, instead of discovering a surprise invoice from three providers at the end of the month.
Computer use — the mode where Jarvis takes control of the pointer and works an app on your behalf — can optionally run on an OpenAI API key you provide, configured under Settings → Computer Use. That is the whole extent of user-supplied keys in Jarvis. It is not a general routing override: it does not change which model answers a normal Cmd+/ prompt, it does not reduce your subscription, and leaving it blank does not disable anything you would otherwise have. Core assistant use never requires a key from you on any plan.
Pricing
