Azure AI services
A group of Azure services, SDKs, and APIs designed to make apps more intelligent, engaging, and discoverable.
3,409 questions
Still got Azure OpenAI Insufficient quota error after reducing existing TPM and RPM
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(96, 15%, 19%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ELP%3C/text%3E%3C/svg%3E)
L P
0
Reputation points
In Azure AI Foundry, I deploy a GPT4.1 with 20K TPM and 20 RPM
I still have 30K quota remain, and I try to deploy a GPT4.1-mini, but I got error "Insufficient quota for selected options"
Thank you!
Update: I have tried delete the GPT4.1, still got the same error when deploying GPT4.1-mini
Azure AI services
{count} votes
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(281.6, 88%, 37%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESA%3C/text%3E%3C/svg%3E)
Saideep Anchuri 6,620 Reputation points Microsoft External Staff2025-05-06T06:43:19.6733333+00:00 Hi L P
It seems that you are encountering an "Insufficient quota" error despite having a remaining quota of 30K.
- Model-Specific Quota Limits: Each model has its own maximum Tokens-Per-Minute (TPM) allocation. For the GPT-4.1 model, the default quota limit is 1M TPM, and for the GPT-4.1-mini, it is also 1M TPM. If the combined TPM of your existing deployments exceeds your total quota, you may not be able to create additional deployments.
- Requests-Per-Minute (RPM): The RPM is also a limiting factor. For GPT-4.1, the RPM is set at a specific ratio to the TPM. If your current deployments are consuming too much of your RPM allocation, it could prevent new deployments.
- Quota Allocation: When you assign TPM to a deployment, it reduces the available quota for that model. If you have already allocated a significant amount of your quota to the GPT-4.1 deployment, it may limit your ability to deploy the GPT-4.1-mini.
If the current quota is not enough, you can request a quota increase for the specific resources needed for the GPT4.1-mini deployment. You can do this by following these steps:
- Go to the Azure portal.
- Select Help + support.
- Choose New support request.
- Provide the necessary information, such as the resource type (GPT4.1-mini), the subscription, and the specific quota you need to increase.
- Submit the request for a quota increase.
Kindly refer below link: quota
Thank You.
-
Saideep Anchuri 6,620 Reputation points Microsoft External Staff
2025-05-07T03:55:36.4933333+00:00 -
L P 0 Reputation points
2025-05-07T04:24:23.2133333+00:00 Hi, I tried to create a support request, but when I clicked the button "Next", the page didn't response, I tried to refresh the page, still don't work
-
L P 0 Reputation points
2025-05-07T04:33:39.3833333+00:00 -
Saideep Anchuri 6,620 Reputation points Microsoft External Staff
2025-05-07T04:49:17.1033333+00:00 Hi L P
It seems you are experiencing issues with creating a support request for your Azure OpenAI quota. If the page is unresponsive when you click "Next," it may be a temporary technical issue. Here are a few steps you can try:
- Clear your browser cache: Sometimes, cached data can cause issues with web applications.
- Try a different browser: If you're using one browser, switch to another to see if the issue persists.
- Check for browser updates: Ensure your browser is up to date, as older versions may have compatibility issues.
- Disable browser extensions: Some extensions can interfere with web pages. Try disabling them temporarily.
In the above screenshot it looks like Your subscription is not eligible for quota increase. please upgrade to pay-as you -go.
Kindly refer below link: quota-increase-for-your-skuThank You.
Sign in to comment
Sign in to answer