mastra-ai/mastra · error · Error

Could not identify provider from model id ${modelId}

Error message

Could not identify provider from model id ${modelId}

What it means

getApiKey parses the model id as 'provider/model'; if either part is missing after splitting on '/', the provider cannot be determined and a plain Error is thrown. The models.dev gateway keys its provider configs by the id prefix, so a malformed id is unusable.

Source

Thrown at packages/core/src/llm/model/gateways/models-dev.ts:313

    // provider default; prefer it when present.
    const perModelApi = config?.modelOverrides?.[modelId]?.api;
    const template = perModelApi ?? config?.url;

    if (!template) {
      return;
    }

    // Check for custom base URL from env vars (explicit override still wins)
    const baseUrlEnvVar = `${providerId.toUpperCase().replace(/-/g, '_')}_BASE_URL`;
    const customBaseUrl = envVars?.[baseUrlEnvVar] || process.env[baseUrlEnvVar];

    return customBaseUrl || interpolateUrlTemplate(template, envVars);
  }

  getApiKey(modelId: string): Promise<string> {
    const [provider, model] = modelId.split('/');
    if (!provider || !model) {
      throw new Error(`Could not identify provider from model id ${modelId}`);
    }
    const config = this.providerConfigs[provider];

    if (!config) {
      throw new Error(`Could not find config for provider ${provider} with model id ${modelId}`);
    }

    const apiKey = resolveApiKeyFromEnv(config.apiKeyEnvVar);

    if (!apiKey) {
      const envVarDisplay = Array.isArray(config.apiKeyEnvVar) ? config.apiKeyEnvVar.join(' or ') : config.apiKeyEnvVar;
      throw new Error(`Could not find API key process.env.${envVarDisplay} for model id ${modelId}`);
    }

    return Promise.resolve(apiKey);
  }

  async resolveLanguageModel({

View on GitHub (pinned to 75dd419e61)

Solutions

1. Pass the full router-style id: 'provider/model' (e.g. 'openai/gpt-4o').
2. Validate the id contains a non-empty provider and model segment before calling.
3. Normalize programmatic id construction to guard against undefined segments.

Example fix

// before
await gw.getApiKey('gpt-4o');
// after
await gw.getApiKey('openai/gpt-4o');

function parseModelId(modelId) {
  const [provider, model] = modelId.split('/');
  if (!provider || !model) throw new Error(`Model id must be 'provider/model', got: ${modelId}`);
  return { provider, model };
}
parseModelId(id); // call before getApiKey

function isValidModelId(id) { return typeof id === 'string' && /^[^/]+\/[^/]+$/.test(id); }

When it happens

Trigger: Calling getApiKey with ids like 'gpt-4o' (no slash), 'openai/' (empty model), '/gpt-4o' (empty provider), or ids containing extra text where the split yields an empty segment.

Common situations: Users passing bare model names from OpenAI SDK habits instead of 'openai/gpt-4o'; building ids programmatically with undefined segments; copying ids from other tools with different id formats.

Related errors

• MODEL_ROUTER_NO_GATEWAY_FOUND
• Could not find config for provider ${provider} with model id
• Invalid model id: ${modelId}
• Could not generate title from input ${JSON.stringify(message
• Expected ${gatewayPrefix}/ in model router ID ${routerId}

AI-assisted analysis of mastra-ai/mastra@75dd419e61 (2026-08-30). Data as JSON: /api/errors/ea5601d4ab4ebcb2. Report an issue: GitHub.