Azure OpenAI Service
An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
Azure OpenAI - 429 - System is experiencing high demand and cannot process your request error
![](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(208, 77%, 30%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EA%3C/text%3E%3C/svg%3E)
Alex
0
Reputation points
We are currently seeing the exception in our OpenAI resource using gpt 4.o in the Australia East region.
Error code: 429 - {"error": {"code": "NoCapacity", "message": "The system is currently experiencing high demand and cannot process your request. Your request exceeds the maximum usage size allowed during peak load. Please retry after 47 seconds. For improved latency reliability, consider switching to Provisioned Throughput."}}
We are using a Standard gpt-4o model with no Provisional Throughput in the Australia East region. This has been occuring between 05/12/2025, 00:07:07.150 and 05/12/2025, 03:27:53.760 with approx. 300 failed requests.
Is there any visibility to when Azure is experiencing high demands and any way we can mitigate this - we do not have capability to use a model deployment in another region.
THanks,
Azure OpenAI Service
{count} vote
-
![](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(83.2, 11%, 18%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAC%3C/text%3E%3C/svg%3E)
André C. Andersen 0 Reputation points2025-12-05T13:36:09.3333333+00:00 We have the same issue for Norway East region. It doesn't even provide any retry window.
{'error': {'code': 'NoCapacity', 'message': 'The system is currently experiencing high demand and cannot process your request. Your request exceeds the maximum usage size allowed during peak load. For improved capacity reliability, consider switching to Provisioned Throughput.'}} -
![](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(176, 16%, 27%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESC%3C/text%3E%3C/svg%3E)
SRILAKSHMI C 10,805 Reputation points Microsoft External Staff Moderator2025-12-05T13:47:52.29+00:00 Hello Alex,
Welcome to Microsoft Q&A and Thank you for reaching out.
I understand that you're experiencing a 429 error with your Azure OpenAI resource in the Australia East region, specifically due to high demand. This can be frustrating, but here are some steps and insights that might help you mitigate the issue:
Understanding the Error
The error message indicates that the system is currently overloaded and unable to process your request. This typically happens during peak load times, and requests can be throttled if they exceed the defined limits.
Suggested Solutions
Implement Exponential Backoff: Introduce a retry mechanism in your application with exponential backoff. This means if you encounter a 429 error, wait some time before retrying the request, increasing the wait time with each subsequent failure.
Monitor and Adjust Usage: Use Azure Monitor to track the request rates and identify any usage patterns or spikes. Ensure the requests are evenly distributed to prevent sudden bursts of traffic.
Check Rate Limits: Make sure you're aware of the rate limits for your subscription. You can review your current limits in the Azure portal under the OpenAI service section, and consider adjusting the
max_tokensandbest_ofparameters for your requests to optimize your usage.Provisioned Throughput: For improved reliability, consider switching to Provisioned Throughput, if feasible. This plan provides a guaranteed amount of capacity, making it less likely to encounter capacity issues during high-demand periods.
Review Quotas and Limits: For a detailed understanding of quotas and best practices, refer to the Azure OpenAI Service quotas and limits documentation.
Ensure you implement retry logic in your application to handle these errors gracefully.
If your demand is consistently high, consider scaling your service in the Azure portal.
Please refer this
- Understanding Rate Limits
- Azure OpenAI Quotas and Limits
- Azure Monitor Documentation
- General Best Practices
I Hope this helps. Do let me know if you have any further queries.
Thank you!
Sign in to comment
Sign in to answer