Persistent 404 Model Not Found Error on New Vertex AI Project (All Regions/Models)

Hello Google Cloud Community,

I am experiencing a persistent `404 Not Found` error when trying to access any Gemini model via the Vertex AI API from my Node.js application hosted on Render. I have already performed extensive troubleshooting and believe this is an internal model provisioning issue with my project.

Project ID:** `metricarmor-v2`

**Error Message:** `GoogleApiError: Publisher Model \`projects/metricarmor-v2/locations/asia-south1/publishers/google/models/gemini-1.0-pro\` was not found or your project does not have access to it.`

**Summary:**

My application is successfully authenticating with a Service Account, but all API calls to generate content fail with a `404`, indicating the models are not found for my project. I have confirmed that using the Vertex AI Studio in the web console **works correctly**. The issue is only with API access from my server.

**Troubleshooting Steps Already Taken:**

1. **Created a Brand New Project:** The current project (`metricarmor-v2`) was created from scratch.

2. **Enabled Billing:** The project is linked to an active and valid billing account.

3. **Enabled All APIs:** I have confirmed that “Vertex AI API”, “YouTube Data API v3”, and “YouTube Analytics API” are all enabled.

4. **Created a Service Account:** I created a new Service Account, granted it the “Vertex AI User” role, and am using its JSON key to authenticate. The authentication is successful.

5. **Tried Multiple Regions:** I have tested with my application’s location set to `us-central1`, `us-east1`, and `asia-south1`. The error is the same in all regions.

6. **Tried Multiple Models:** I have tried calling `gemini-1.0-pro`, `gemini-1.5-flash-001`, and `gemini-pro-vision`. All return the same `404 Model Not Found` error.

7. **Performed a “First Use” Action:** I have successfully sent a prompt from the Vertex AI Studio in the console to initialize the service.

This appears to be a backend provisioning issue where the generative models are not being made available to my project’s API access, despite all settings being correct in the console.

Could a Google engineer please investigate the internal permissions and model provisioning for my project `metricarmor-v2`?

Thank you for your help.