----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Function Calling]() * [Outbound Sales Example]() * [Web]() * [Create Phone Call]() -----------------------------

### Answer It seems like you're experiencing intermittent issues with function calls in Vapi, which can be influenced by various factors such as the configuration of your assistant, the complexity of the function being called, or even the chosen language model and transcriber. While the official documentation does not directly address intermittent issues with function calls to specific services like Make.com, there are a few general areas you might consider reviewing to ensure consistency: 1. Parameter Names in Function Calls: Make sure the parameter names in your function calls match those expected by your server endpoint. There have been instances where parameter names got changed in the server message POST request, which could lead to inconsistent behavior [(1)]() . 2. Choice of Language Model or Transcriber: Different language models (LLMs) and transcribers might handle data differently, which could impact the consistency of function calls. It's worth experimenting with different LLMs or transcriber settings if you're noticing inconsistencies potentially related to how data is being processed or understood [(2)]() . 3. Server Configuration: Ensure your server URL and the function call's setup are correctly configured. The server URL can be set at various levels, and the specificity of the URL might affect the behavior of your function calls. Review the hierarchy of server URL settings to ensure they are accurately pointing to your intended endpoint [(3)]() . 4. Reviewing Documentation and Examples: Revisiting the documentation, especially sections related to function calling [(4)]() and any re

levant examples or guides [(5)]() , might offer insights or highlight a missing or misconfigured step in your setup. If the problem persists and seems to deviate from common configurations or expected behaviors documented, it could be beneficial to explore community forums or reach out to the Vapi support team directly for more targeted assistance, as intermittent issues can stem from nuanced configurations or specific use cases not broadly covered in the documentation.

Got it. Tagging @User for additional help.

Currently, Groq doesn't support function calls via streaming due to which function calls don't work. So, can you provide me some call_ids where you used GPT-4?