Hello Vapi team, Is there any way to configure a model for call analysis for my assistant? (I coudn't find any in the API reference) Or at least to make sure that the model stays the same as it's important for prompt evaluations? We have created and tested the call analysis configuration of our assistant against Anthropic’s Claude Sonnet model. But when I actually checked the data of one finished call, I was surprised to see that a totally different model was used: Call ID: 5e1a7e32-90ca-4976-a44f-01c7d696eab9

{
      "cost": 0.00053625,
      "type": "analysis",
      "model": {
        "model": "gemini-2.5-flash-preview-04-17",
        "provider": "google"
      },
      "analysisType": "structuredData",
      "promptTokens": 2047,
      "completionTokens": 382
    },

According to the official Vapi documentation, Claude or ChatGPT (as a fallback) must be used for call analysis. Instead a less capable (IMHO) model was chosen for my call.

Currently, Vapi does not provide a way to configure or lock the model used for call analysis. According to the documentation, call analysis is typically performed using Anthropic’s Claude Sonnet, with GPT-4o as a fallback, and there is no user-facing option to select or guarantee the model. If you are seeing a different model (like Gemini) in your call records, this is not documented behavior, and you may want to reach out to Vapi support for clarification or to report this unexpected result. Source: - [Call analysis documentation](https://docs.vapi.ai/assistants/call-analysis)

Okay, I just stumbled up the announcement (somehow I missed it during the text search): https://discord.com/channels/1211482211119796234/1211483332496465962/1377873568876068915 Is it possible to revert it back to sonnet for us?

Hey Vadym, I have disabled the google gemini analysis model for your organization. Try 2-3 calls and let me know how it goes.

Hey Shubham, thank you so much. It works well!

Marking this ticket as Solved ✅