Unable to increase GPU quota - GPUS_ALL_REGIONS stuck at 0, console and CLI both rejected

Google Cloud Compute Infrastructure
1

Hi,

I am requesting help with a GPU quota increase for my project.

Requested Quotas

Quota Current Requested Scope
GPUS_ALL_REGIONS 0 1 Global
NVIDIA_L4_GPUS 0 1 us-central1
NVIDIA_TESLA_T4_GPUS 0 1 us-central1

Project Description

I am working on a Korean TTS (Text-to-Speech) model fine-tuning project. Using Vertex AI Colab Enterprise with PyTorch, I only need 1 GPU (T4 or L4) for
small-scale training. The instance will auto-shutdown after 1 hour idle, with estimated usage of 50-100 GPU hours/month (~$35-$140).

Current Issue

The GPUS_ALL_REGIONS quota is set to 0, making it impossible to create any GPU instance.

  • Console: The edit panel says “Enter a new quota value between 0 and 0” — no increase is possible.
  • CLI: gcloud alpha quotas preferences create is auto-rejected with “We cannot grant the preferred quota at this moment.”
  • Colab Enterprise: Runtime creation fails with “Quota GPUS_ALL_REGIONS exceeded. Limit: 0.0 globally.”

Billing is active, Compute Engine API and Vertex AI API are both enabled. I only need a minimal allocation of 1 GPU to proceed with my project. Any help would be
greatly appreciated.

Thank you.