Reference
Arcane.ollama.running()
Lists the provider running-model envelope.
This focused page is derived from the mechanically checked full member inventory.
Syntax
Arcane.ollama.running()
Parameters
None
Return value
Promise<OllamaRunningModelsResponse>
Description
Lists the provider running-model envelope.
Overview
Arcane.ollama.running() reads Ollama's currently loaded-model envelope. Use
it for diagnostics such as residency display. It is a momentary provider
snapshot, not proof that capacity or admission will remain available.
Parameters
The method takes no arguments.
Return value
It resolves to the bounded provider-native /api/ps JSON, usually with a
models array. Arcane does not normalize its nested provider fields.
Availability
This is a desktop Core diagnostic read requiring ai.models.read, admitted
only to Settings, Shell, and Terminal. Android and browser previews do not
expose it.
Errors and recovery
An unavailable provider, failed HTTP request, invalid JSON, or oversized response rejects. Refresh local service status before retrying. An empty provider array is a valid no-model-loaded state, not an error.
Streaming, cancellation, and events
This method does not stream or emit events and has no signal. Core's provider timeout is ten seconds. Models can load or unload immediately after resolution.
Example
const running = await globalThis.Arcane.ollama.running();
for (const model of running.models ?? []) {
console.info(model.name ?? model.model ?? 'Unnamed loaded model');
}