continuedev/continue · warning

Failed to fetch models for ${provider}: ${error?.message ??

Error message

Failed to fetch models for ${provider}: ${error?.message ?? error}

What it means

A generic wrapper thrown by fetchProviderModelsViaListModels when constructing an LLM client for a provider and calling llm.listModels() fails for any reason; the original error message is embedded with the provider name. It aggregates auth, URL, and connectivity failures for arbitrary providers.

Source

Thrown at core/llm/fetchModels.ts:235

async function fetchProviderModelsViaListModels(
  provider: string,
  apiKey?: string,
  apiBase?: string,
): Promise<FetchedModel[]> {
  try {
    const cls = LLMClasses.find((llm) => llm.providerName === provider);
    const defaultApiBase = cls?.defaultOptions?.apiBase;

    const llm = llmFromProviderAndOptions(provider, {
      apiKey,
      apiBase: apiBase || defaultApiBase,
      model: "",
    });
    const modelIds = await llm.listModels();
    return modelIds.map((id) => ({ name: id }));
  } catch (error: any) {
    throw new Error(
      `Failed to fetch models for ${provider}: ${error?.message ?? error}`,
    );
  }
}

export async function fetchModels(
  provider: string,
  apiKey?: string,
  apiBase?: string,
): Promise<FetchedModel[]> {
  switch (provider) {
    case "ollama":
      return fetchOllamaModels();
    case "openrouter":
      return fetchOpenRouterModels();
    case "anthropic":
      return fetchAnthropicModels(apiKey);
    case "gemini":

View on GitHub (pinned to 5522c6f44c)

Solutions

1. Read the embedded inner message — it identifies the real cause (connection refused vs 401 vs invalid URL)
2. Verify the provider's apiBase is correct and the server is running (curl the base URL)
3. Check the provider's API key/credentials
4. Retry transient network failures with backoff

Example fix

// before
{ provider: "openai", apiBase: "http://localhost:1234" } // server down
# start the server, then:
// after
{ provider: "openai", apiBase: "http://localhost:1234" } // curl http://localhost:1234/v1/models now returns JSON

const health = await fetch(apiBase.replaceAll(/\/$/, '') + '/models'); if (!health.ok) disableProviderModelListing();

catch (e) { if (e.message.startsWith('Failed to fetch models for')) { inspectInnerMessageAndFix(e.message); } else throw e; }

When it happens

Trigger: Any provider whose listModels call fails: wrong apiBase, unreachable server, invalid API key, or a provider constructor throwing on model: "".

Common situations: Misconfigured apiBase for LM Studio/vLLM/Ollama-style local servers; server not running; custom provider with a bad fetch implementation; auth failures for hosted providers.

Related errors

• Failed to fetch Ollama library: ${response.status}
• Failed to fetch OpenRouter models: ${response.status}
• Failed to fetch Anthropic models: ${response.status}
• Failed to fetch Gemini models: ${response.status}
• Not enough context available to include the system message,

AI-assisted analysis of continuedev/continue@5522c6f44c (2026-08-27). Data as JSON: /api/errors/c4f3da64f6081653. Report an issue: GitHub.