you check why the call cost became so high in our vapi account for the past 2 days? for example call IDs are 0199f8b1-6678-7cc3-835d-b3fd5a4afe49 0199f66f-f5f9-7cc3-b7f2-49449b01953b

Thank you for providing the call IDs. To help investigate the high call costs, could you please confirm if there were any unusual call durations, destinations, or changes in your Vapi configuration for these calls in the past 2 days? This will help narrow down the possible causes. Sources: - [Debugging voice agents](https://docs.vapi.ai/debugging) - [How to Report Issues Effectively](https://docs.vapi.ai/issue-reporting)

It was a usual less than 6 min call. we are in debugging mode of the assistant, so its expected that call can get a little lengthier

We reviewed the call logs and it looks like LLM is the main source of your costs:

{
      "cost": 1.080786,
      "type": "model",
      "model": {
        "model": "claude-opus-4-20250514",
        "provider": "anthropic"
      },
      "promptTokens": 63667,
      "completionTokens": 1063
    },

This is mainly due to large responses from the tool calls. If you would like to reduce your costs, please shorten your tool call results to only include the bare essentials for your request. Returning large amounts of data multiple times in a call will incur large fees