[VertexAI.ClientError]: got status: 404 Not Found. {"error":{"code":404,"message":"Publisher Model projects/angelic-bee-418405/locations/us-central1/publishers/google/models/gemini-1.5-flash was not found or your project does not have access to it. We have over 1000 assistants using query tools with 1.5-flash so we are unable to easily update this to 2.0 Support please investigate

Action Step: Check if the Gemini 1.5-flash model is still available in your Google Cloud project and that your service account has access. In your Vapi configuration, ensure the model and provider are set exactly as shown below:

json
{
  "provider": "google",
  "models": ["gemini-1.5-flash"]
}

If the model is missing or access has changed, you may need to contact Google Cloud support to restore access or update your deployments. Reference: See [Gemini model support](https://docs.vapi.ai/changelog/2024/10/29) and [model configuration](https://docs.vapi.ai/changelog/2025/3/6). Sources: - [Gemini model support and credential management](https://docs.vapi.ai/changelog/2024/10/29) - [Query Tool and model configuration](https://docs.vapi.ai/changelog/2025/3/6)

I'm getting this exact same error

Hi Louis, Based on the changelog, there has been a significant update regarding Gemini models: The

gemini-2.0-flash-lite

model was introduced as a new option for

Assistant.model[provider="google"].model

. This model offers: - Reduced latency - Lower cost - 1 million token context window [1](https://docs.vapi.ai/changelog/2025/3/15) Since you're encountering a 404 error with

gemini-1.5-flash

, this suggests the model may have been deprecated or is no longer accessible. For bulk updates across your 1000+ assistants, you would need to: 1. Update the model configuration to use

gemini-2.0-flash-lite

2. Test the new model configuration 3. Plan a systematic migration However, given the scale of your deployment, this should be coordinated with support to ensure a smooth transition and minimize disruption to your services.