New Project - "404 Model Not Found" for all Gemini Models in Vertex AI

Google Cloud Build with AI Generative AI & Foundational Models
1

Hello,

I have a new Google Cloud project (ID: ai-social-media-report) with a newly activated billing account. I am trying to use a Gemini vision model (e.g., “gemini-pro-vision”) in a Cloud Function.

The Vertex AI API is enabled, and my function’s service account has all the necessary roles (Vertex AI User, Service Usage Consumer, etc.).

However, every attempt to call a model results in a “404 Publisher Model not found” error. I have confirmed this in multiple regions.

A diagnostic script that lists available models (PII Removed by Staff) confirms that no Gemini models are being returned for my project.

Since I am on the free support plan, I cannot open a technical case. Can anyone please look into my project to see why the Gemini models are not being provisioned or made accessible?

Thank you.

3

I am experiencing the same 404 model not found as well, all the sudden yesterday (9/25) at around 2:38PM EST.
Gemini API Requests Incorrectly Routed to Old Model Endpoint
My Gemini API calls are failing with a 404 error for the model gemini-1.5-pro-002, even though my code explicitly requests gemini-1.5-pro-latest.

This issue started suddenly on Thursday, September 25, 2025.

The full error message is: 404 POST ... Publisher Model 'projects/generativelanguage-ga/locations/europe-west4/publishers/google/models/gemini-1.5-pro-002' was not found...

This is an account-level issue that I have confirmed by:

  1. Using a minimal test script in a clean Google Colab environment.

  2. Generating a brand new API key in a brand new project.

The error persists in all scenarios. It appears my account’s API requests are being incorrectly routed to the europe-west4 location, which is causing the failure.

4

I do more testing, and believe found the resolutions. the 1.5 may have been retired and no longer enabled since yesterday. I changed my code from the original: model = genai.GenerativeModel(‘gemini-1.5-pro-latest’)

To: model = genai.GenerativeModel(‘gemini-2.5-pro-latest’)

5

I updated to the 2.5 version and getting this erorr still code: 404,
message: ‘models/gemini-2.5-pro-latest is not found for API version v1beta, or is not supported for generateContent. Call ListModels to see the list of available models and their supported methods.’,
status: ‘NOT_FOUND’
}
}

6

appears changing this to gemini-2.5-flash is now working but its running slower.

7

guys they deprecated 1.5 models, if you’re still getting errors from 2.5 models, that’s your real problem.

Aaron, the error would be v1beta api. can you use normal genai-sdk? (like up-to-date api?)

8

wow, i’m experiencing the same issue on project (PII Remove by Staff) from the southamerica-east1 region. list_models() shows the models, but any call to generateContent returns a 404 Not Found error. This is a critical blocker!!