{{template "shellTop" "models"}}

Resident model

swapping requires step-up — assert a passkey first. The old model is unloaded before the new one is loaded (one model fits the iGPU at a time), so turns during the load are refused and fall back to the classifier.

a swap is not remembered. Nothing writes it down, so the next restart of the daemon — including the one mavupdate does — comes back on phraser.model_path from the config. Make it stick by editing that.

{{if .Msg}}
{{.Msg}}
{{end}} {{if .Err}}
{{.Err}}
{{end}} {{if .Off}}

swap not configured

this core has no phraser.swap_models allowlist, so there is nothing to swap to. Add the gguf paths you allow to deploy/mavend.json and restart once.

{{else}}

loaded now

model{{.Status.Model}}
file{{.Status.ModelPath}}
server{{.Status.BaseURL}}
n_ctx{{.Status.NCtx}}
n_gpu_layers{{.Status.NGpuLayers}}

the model name is what llama-server reports for itself, not what the config says it should be.

allowed models {{len .Status.Swappable}}

{{if .Status.Swappable}}
{{range .Status.Swappable}}{{end}}
file
{{.}}
{{else}}
no models allowlisted
{{end}}
{{end}} {{template "shellBottom"}}