Reference
Arcane.ollama
Admitted native Ollama inference, inventory, policy, and model-management operations.
This focused page is derived from the mechanically checked full member inventory.
Syntax
Arcane.ollama
Member kind
Namespace
Description
Admitted native Ollama inference, inventory, policy, and model-management operations.
Overview
Arcane.ollama exposes admitted Ollama inference, raw diagnostic inventory,
managed selection and settings, and policy-controlled model management. Raw
provider inventory is not an application's model-admission catalog. Streaming
calls can deliver correlated ollama.chunk events or an onChunk callback.
Availability and security
Inference requires ai.inference; diagnostic inventory requires
ai.models.read and is limited to Settings, Shell, and Terminal. Mutations use
ai.models.manage or Settings-only ai.settings.manage, frequently behind an
exclusive mutation boundary. Android user-managed loopback admission exposes
chat only and does not grant managed model lifecycle authority.
Example
const access = await globalThis.Arcane.capabilities.list();
const canUseOllamaChat = access.methods.includes('ollama.chat');
console.log('Native Ollama chat admitted:', canUseOllamaChat);
Members
Arcane.ollama.version()— Gets the managed Ollama version for Settings, Terminal, and Shell diagnostics.Arcane.ollama.models()— Lists the raw provider inventory envelope; it is not the application admission catalog.Arcane.ollama.list()— Alias of the raw models() inventory.Arcane.ollama.running()— Lists the provider running-model envelope.Arcane.ollama.show()— Gets model metadata and configuration.Arcane.ollama.generate()— Generates text after Core re-admits the exact requested model; optionally streams chunks.Arcane.ollama.chat()— Runs native Ollama chat after the host re-admits the exact requested model, optionally streamed. An admitted Android host proxies only this inference operation to its user-managed fixed loopback service and exposes no model mutation or lifecycle operation.Arcane.ollama.embed()— Creates embeddings after Core re-admits the exact requested model.Arcane.ollama.pull()— Direct application pull fails closed; Arcane's managed workflow owns integrity-bound preflight and admission.Arcane.ollama.push()— Pushes only a model authorized by verified application policy.Arcane.ollama.create()— Creates a package-owned verified alias only after from, normalized system, and parameters match policy and files/adapters are absent; other admitted fields remain provider-native.Arcane.ollama.copy()— Raw application alias copy is denied; managed selection owns alias mutation.Arcane.ollama.delete()— Deletes only a model authorized by verified application policy.Arcane.ollama.selection()— Gets Arcane's managed model preference and state.Arcane.ollama.select()— Selects a managed size preference and reconciles the model.Arcane.ollama.settings()— Gets Ollama runtime settings.Arcane.ollama.saveSettings()— Saves only runtime-owned default-model, boot-load, keep-alive, and context settings.Arcane.ollama.createBrain()— Creates an Arcane brain model.Arcane.ollama.serviceSettings()— Gets managed Ollama service settings.Arcane.ollama.saveServiceSettings()— Saves managed service settings.