Hello,
I am unable to access any Imagen publisher model in a newly created Google Cloud / Vertex AI project.
Environment:
- Region: us-central1
- Vertex AI API: enabled
- Billing account: active
- Model access organization policy (
vertexai.allowedModels): not enforced - Vertex AI service agent: created and granted
roles/aiplatform.serviceAgent - OAuth authentication using a service account: successful
- Vertex AI TTS request: successful
- Gemini image generation (Nano Banana Pro) in the Google Cloud console: successful
However, all of the following Imagen models return the same 404 error when tested directly in Vertex AI Studio, without using n8n or any external workflow:
imagen-4.0-fast-generate-001imagen-4.0-generate-001imagen-4.0-ultra-generate-001imagen-3.0-capability-001
Example endpoint:
https://us-central1-aiplatform.googleapis.com/v1/projects/[REDACTED-PROJECT-ID]/locations/us-central1/publishers/google/models/imagen-4.0-fast-generate-001:predict
Error:
Publisher model ... was not found or your project does not have access to it. Ensure you are using a valid model name and that the model is available in the specified region. Status: NOT_FOUND (404).
Additional verification:
The exact same preflight workflow, model name, API version (v1), region (us-central1), and :predict request succeeded in my previous Google Cloud project on July 27, 2026. In the new project, credential validation, OAuth token creation, and Vertex AI TTS all succeed, but the Imagen request alone returns 404.
I also recreated the Vertex AI service identity and verified its service-agent role, but the issue remains.
Could a Googler please confirm whether this new project is missing Imagen publisher-model access or requires an additional backend entitlement/allowlist activation? I can provide the full project ID and diagnostic report privately if required.
Thank you.