Cannot access Gemini text models via API Key (404 Not Found), but embedding models work

Google Cloud Build with AI Generative AI & Foundational Models
1

models/gemini-pro is not found for API version v1beta, or is not supported for generateContent.

Hello,

My application is consistently receiving a 404 Not Found error when trying to call the ‘gemini-pro’ model using the Generative Language API (generativelanguage.googleapis.com).

I have diagnosed the root cause. When I call the v1beta/models endpoint to list available models for my API key, the ONLY model returned is models/embedding-gecko-001 or embedding-001. No text generation models like gemini-pro or gemini-1.5-flash are listed as available for my account.

This confirms the issue is with my account’s provisioning. I have already verified the following:

  1. Billing Account: My account is a full, paid account, upgraded from the free trial.
  2. API Enablement: I have manually verified that the “Generative Language API” is enabled in my Google Cloud project.
  3. Network Connectivity: Test-NetConnection from my server (Public IP: 180.250.xx.xx) to the API endpoint is successful (TcpTestSucceeded: True).
  4. API Key Configuration: The API key is correctly configured and explicitly restricted to the “Generative Language API”.

Please escalate this issue to resolve the provisioning of my account so that I can access the standard text generation models (e.g., gemini-pro) via my API key.

Thank you.