Unable to see VertexAI quota for gemini-2.0-flash-001

[thisisvaze]

Hi everyone,

I’m currently working with Vertex AI and trying to use the ⁠gemini-2.0-flash-001 model.

However, I noticed that I’m unable to see the quota for this model in my Google Cloud project.
I’ve checked the Quotas page in the Google Cloud Console, but the specific quota for ⁠gemini-2.0-flash-001 doesn’t appear. I’m not sure if this is a permissions issue, a regional availability problem, or if additional steps are required to enable this model in my project.
Has anyone else experienced this? If so, how did you resolve it?

[MarvinLlamas]

Hi @thisisvaze,

Welcome to Google Cloud Community!

It seems like you're dealing with the issue of not being able to view the quota for the gemini-2.0-flash-001 model in your Google Cloud project's Vertex AI environment.

Here are the potential ways that might help with your use case:

• Regional Availability and Model Access: You might want to double-check the region you're using for your Vertex AI resources, as the gemini-2.0-flash-001 model may not be available everywhere. If needed, move your resources to a supported region by creating new notebooks, transferring your data, and retraining models.

• Permissions and Roles: Make sure you have the necessary IAM permissions to view quotas for Vertex AI models. Verify the roles assigned to your account in the IAM section of the Google Cloud Console.

• Quota Display Delay or Caching Issues: Sometimes quota information takes time to appear in the Google Cloud Console after enabling a new service or model. Caching issues can also prevent up-to-date information from displaying. To resolve this, wait a few hours and refresh the Quotas page, clear your browser's cache and cookies, try incognito mode, or use a different browser.

• Incorrect Quota Search Filters: You might need to adjust your filter criteria when searching for the quota in the Cloud Console. Try clearing any filters you've applied, using broad search terms like "Gemini" or "AI Platform," and setting the "Region" selector to "All regions" or your specific region.

• IAM Permission: Ensure that your IAM role has sufficient permissions to view quotas. Verify that your role includes `serviceusage.quotas.get` and `serviceusage.quotas.list` permissions. The "Owner" or "Editor" role typically includes these, but a more restricted role may require adjustments. If necessary, grant the required permissions to your account or the service account you're using.
• Service Not Enabled: Ensure that your core AI Platform service is enabled, even if you're primarily using Vertex AI. Go to the API Library in your Google Cloud Console and search for "AI Platform Training." If it's not enabled, activate it, as it's often required for your Vertex AI models.

You can refer to the following documentation on Google Cloud's Vertex AI for detailed information on supported models and learning resources:

• Google models
• Supported models

Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.

[RayL]

I am seeing this as well. I think this is some serious issue on the Vertex AI / Gemini side

[thisisvaze · Accepted solution]

@MarvinLlamas @RayL actually I had to dig through the internet to find the correct answer. I wish it was more easily accessible information from Google Cloud.

Gemini 2.0 models have dynamic quota and hence don't appear on quota information list.

https://cloud.google.com/vertex-ai/generative-ai/docs/dsq