Arcane OS Docs Development only

Reference

Arcane.ai.providerModels()

Queries models available to the configured OpenAI account.

  • Reference

This focused page is derived from the mechanically checked full member inventory.

Syntax

Arcane.ai.providerModels()

Parameters

None

Return value

Promise<AIProviderModelCatalog>

Description

Queries models available to the configured OpenAI account.

Overview

Arcane.ai.providerModels() lists OpenAI model identifiers accessible to the stored credential. Use it to populate the Settings model control, not to list local Ollama models.

Parameters

The method takes no arguments.

Return value

It resolves to {provider: "openai", models}, where models is a sorted array of bounded account-accessible identifiers.

Availability

This is a desktop Core, Settings-only method requiring ai.settings.manage. It performs one protected OpenAI model-list network request and requires a stored credential even when Ollama is currently selected. Android does not project it.

Errors and recovery

OPENAI_NOT_CONFIGURED requires the user to save a credential. Provider unavailable, request-failed, response-too-large, or invalid-response errors are not evidence that a previously saved model should be deleted; keep the current choice and offer a later retry.

Streaming, cancellation, and events

This is one non-streaming network read with no method event or per-call signal. The wrapper timeout is 130 seconds. Avoid issuing a request for every render or keystroke; refresh it as one owned Settings task.

Example

const catalog = await globalThis.Arcane.ai.providerModels();

for (const modelId of catalog.models) {
    console.info(modelId);
}

Reference group

AI and Ollama

Repository and reviewed source access