Skip to main content

Models API

The /v1/models endpoint is your window into the available AI models. By calling this route, you can discover which models your application has access to—ensuring you select the right one for your task, whether it’s chat, completions, or specialized processing.

Pro Tip: Use /v1/models at startup to dynamically fetch the model list, so your app can adapt to newly added or deprecated models without code changes.

How It Works​

When you send a request, the API responds with a list of model objects, each describing:

  • id: The unique identifier for the model (e.g., gpt-5, gpt-4-mini).
  • object: The type of resource returned (always model).
  • owned_by: The provider or owner of the model.

This lets you:

  • Quickly enumerate all available models.
  • Validate that a model ID exists before sending requests.
  • Adjust your app to feature flags (e.g., only enable pro features if enterprise+ models are available).

Note: Not all accounts will see the same list—availability depends on your plan, permissions, and enterprise codes.