Hi team — requesting a DSQ allocation increase for the gemini-2.5-flash-image-ga model on Vertex AI.
Project: outfitk-app
Service: aiplatform.googleapis.com
model : gemini-2.5-flash-image
Region: us-central1
Billing: enabled
Current state: No editable Type=Quota row exposed in the Cloud Console quotas page for this model — only Type=System limit rows are visible. The Cases tab is blocked on free support tier (“no permission to file tech-related support cases”). The gcloud alpha quotas preferences create command also failed with ADC/billing-project errors despite enabling cloudquotas.googleapis.com.
Requested change: Increase DSQ allocation for gemini-2.5-flash-image in us-central1 to support 600 RPM (or whichever Tier 2+ allocation is appropriate for our profile below).
Use case: Outfitk (outfitk com) is a B2B virtual try-on product launching to public traffic. Expected load is 500–2,000 daily active users in month 1, with 5–10% concurrent peaks. Each session triggers 2–5 image-generation calls.
Application-level rate limiting: 2 generations per user per minute, admin-configurable, already deployed.
Stack: Vercel (Next.js 15), Supabase, Cloudflare R2.
Auth: service account (GCP_VERTEX_SA_JSON).