Assistant
Three speeds, or bring your own key
A built-in AI chat with three modes — Aurora, Nebula and Pulsar — served through luma's own proxy, or pointed at your own provider account if you would rather pay for it directly.
What's in it
-
Aurora, Nebula, Pulsar
Fast, smarter, and smartest. If the server operator has configured a shared key it works out of the box with a fair per-user allowance.
-
Weighted allowance
Nebula counts five times and Pulsar twenty, so one expensive reply is not billed like a cheap one.
-
Bring your own key
Use your own Anthropic, OpenAI, Mistral or Google key and talk to those providers straight from your device.
-
Keys stay encrypted
Provider keys are encrypted at rest in their own key file, separate from the password vault.
Luma 1.1
Every module ships in every build
There is no paid tier hiding any of this. Grab the latest build straight from CI.