Struggling to find/increase gemini-3.5-flash TPM/RPM Quota Metrics

Google Cloud Build with AI Agents
, , ,
1

Hello Google Cloud Community,

We are running an AI-powered platform in production. Our primary workload and database are deployed in the europe-west1 region, and our core LLM application logic is explicitly configured to use the gemini-3.5-flash model via the GCP Vertex AI API (not Google AI Studio).

As our enterprise user base and concurrent document-processing workloads are growing, we are starting to hit rate limits (HTTP 429 RESOURCE_EXHAUSTED), and we want to proactively scale up our capacity.

However, we are having trouble identifying the exact quota rows inside the Google Cloud Console “Quotas & System Limits” dashboard, and we do not currently have a paid Technical Support plan to open technical cases.

We would appreciate advice from the community and Google experts on the following:

  1. Quota Mapping: Under which specific GCP Quota metrics (and dimensions) are the Requests Per Minute (RPM) and Tokens Per Minute (TPM) for the gemini-3.5-flash model tracked in europe-west1/global? Are they shared with gemini-1.5-flash / generic “Generate content” metrics, or are there dedicated limits we should search for?

  2. Location/Region: Since we are using “global” model routing (GOOGLE_CLOUD_LOCATION=global) but our app is hosted in europe-west1, which region dimensions should we submit the quota increases for?

  3. Recommended Limits: We would like to request:

- Tokens Per Minute (TPM): 40,000,000 (40 Million)

- Requests Per Minute (RPM): 3,000

- Generate content concurrent requests: 100

Do you guys have an experience with requesting such limit increases for Gemini 3.5 Flash, and what is the best way to get them approved quickly for a growing startup?

Thank you in advance for your guidance!

1 Like