Project ID: lenzi-5012..
Project number: 155422317…
Model: gemini-embedding-2
Client: google-genai Python SDK 2.7.0
Auth: Gemini API key, not Vertex AI
SDK base URL: https://generativelanguage.googleapis.com/
Google Cloud Console > Gemini API quota shows:
generativelanguage.googleapis.com/embed_content_paid_tier_requests
model=gemini-embedding-2
limit=3000 RPM
But API calls fail with:
429 RESOURCE_EXHAUSTED
Quota exceeded for aiplatform.googleapis.com/global_embed_content_requests_per_minute_per_base_model
base model: gemini-embedding-2
Cloud Service Usage shows that hidden/enforced aiplatform quota is only 50 RPM:
aiplatform.googleapis.com/global_embed_content_requests_per_minute_per_base_model
base_model=gemini-embedding-2
effectiveLimit=50
Please resync/fix the project so Gemini API embed_content uses the correct Gemini API paid tier quota.