I'm experiencing oddly slow response times and timeouts on my OpenAI services. Please find the information below:
Unable to access your assistants
Timeout of 60000ms exceeded.
Trace ID : a58a1fa0-292b-42d9-a0f7-e47e99b5e211
Client request ID : 8c9f57b4-6b7d-4348-8d2e-8c8f186b5b8a
Hi
Benjamin Fistein
The timeout message you're experiencing indicates that the request to your OpenAI service is exceeding the specified timeout limit of 60000 milliseconds (60 seconds). This can happen for various reasons, including server-side issues or client-side configurations.
The default timeout for requests to model serving endpoints can vary. For CPU or GPU serving endpoints, the default is 120 seconds, while for foundation model serving endpoints, it is 297 seconds. If your requests are timing out at 60 seconds, it might be due to a configuration issue on the client-side.
If you're using a third-party client API, ensure that the timeout settings are configured correctly. For example, with the OpenAI client, you can adjust the
timeout
parameter to allow for longer request times.
Check the service logs for any errors or additional information that might help identify the cause of the slow response times.
I hope these helps you Thank you!
@Anonymous
We do have the same issue, since yesterday. The problem is, that in our Subscription we have an Azure OpenAi gpt-4o deployment in the datazone region with 2024-05-13 model version. Since yesterday at 09:30 AM CEST the Assistants API in this subscription is not available. We have a lot of user who are using this endpoint.
We cannot call the Assistants api with the Python Azure OpenAI library and also the Assistants do not show up in the Azure Open AI Service Web Playground in Azure Ai Foundry.
The issue does only occur in this subscription, we have another subscription, with the same configuration where the API is available.
I'm also experiencing the same problem today (wed 19.3.2025 10:00UTC) in swedencentral with gpt-4o, 2024-11-20, Assistants. Can't access the assistants either from AI Foundry nor using the API.
Interestingly, this only affects the gpt deployment in one of my resource groups, but a resource group I created this Monday can access its deployment's Assistants fine.
From the API:
"The server had an error processing your request. Sorry about that! You can retry your request, or contact us through our help center at oai-assistants@microsoft.com if you keep seeing this error. (Please include the request ID 43a22a92-26bf-4b9e-bc13-c546e787228c in your email.)"
From AI Foundry:
Unable to access your assistants
Timeout of 60000ms exceeded.
Trace ID : 87f574c4-1587-46c5-a31a-0823d6df4754
Client request ID : 583ae572-6323-4aa5-9e03-ed415ab1d0cc
Unable to access your assistants
Timeout of 60000ms exceeded.
Trace ID : c9d664de-b76e-47e2-9b20-163756c1f6fa
Client request ID : 1955340b-f108-494a-bc7a-3fa492942761
