----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Provider Keys]() * [Web Calling]() * [Custom LLM]() * [Dashboard]() -----------------------------

### Answer If the OpenAI models are not appearing correctly in the assistant options despite having your OpenAI API key set under providers, it might be due to a few reasons: 1. **Issue with API Key**: - Ensure that the API key you provided is correct and valid. Invalid keys may not allow access to the models. You can check for errors in the network tab as mentioned in the Discord forum [(1)]() . 2. **Supported Models**: - Vapi supports using any OpenAI-compatible endpoint. Ensure you are specifying the correct parameters when setting up the assistant. For example, your assistant setup might look like this:

json
    {
      "name": "My Assistant",
      "model": {
        "provider": "openrouter",
        "model": "gpt-4",
        "messages": [
          {
            "role": "system",
            "content": "You are an assistant."
          }
        ],
        "temperature": 0.7
      }
    }

[(2)]() , [(3)]() 3. **Custom LLMs**: - If you are using custom LLMs or different providers, make sure they are correctly set up and the APIs endpoints are reachable. More details can be found in documents about [Custom LLMs]() [(2)]() . Ensure you have followed the setup instructions properly and double-check the API keys and configurations.

you should see gpt-4o as well

@Sahil I assume the built in functions not working is a bug?