Consult Mode — when does the cloud LLM kick in?
2 min
Only when 3 local SLM iterations don't converge AND the request explicitly opts in.
Local Failover (v1.6 — formerly called Cloud Consult / Consult Mode) is OFF by default. When ON, the Runtime path's local SLM call gets retried against your configured BYOK cloud LLM if the local call fails. Two gates have to flip:
1. Settings → AI → AI Model for Runtime is set to Local (not Cloud — Local Failover is N/A in Cloud mode). 2. The Local Failover card on the Runtime tab has a cloud provider selected + the toggle is ON.
When escalation fires, your BYOK key (encrypted in the local keystore) is used to authenticate. Anonymised problem + plan get sent — never your description, project files, or PII.
The sovereignty boundary: Avery's Central Server is NOT in this path. The desktop talks directly to your chosen cloud provider with your key. NB — Build is a separate concern (always cloud, configured in AI Model for Build); Local Failover only affects Runtime.
Live recipes need the desktop
This article is a static preview. The in-app Help sidecar inside Avery NXR can fire each step against your live project — install the desktop to use it interactively.