- Package Name: azure-ai-projects
- Package Version: 2.6.0
- Operating System: Mac OS
- Python Version: 3.14
Describe the bug
GPT-6-astra not working when using the project client.
To Reproduce
Steps to reproduce the behavior:
- Run the following code (with a gpt-6-astra deployment, change deployment name if different)
from azure.ai.projects import AIProjectClient
from azure.identity import DefaultAzureCredential
project_client = AIProjectClient(
endpoint="https://<foundry-instance>.services.ai.azure.com/api/projects/<project>",
credential=DefaultAzureCredential()
)
openai_client = project_client.get_openai_client()
openai_client.responses.create(
input="hello",
model="gpt-6-astra" # or whatever the deployment is named
)
- I get an error as below:
InternalServerError: Error code: 500 - {'error': {'message': 'The server had an error processing your request. Sorry about that! You can retry your request, or contact us through an Azure support request at: https://go.microsoft.com/fwlink/?linkid=2213926 if you keep seeing this error. (Please include the request ID <redacted> in your email.)', 'type': 'server_error', 'param': None, 'code': 'server_error', 'request_id': '<redacted>'}}
Expected behavior
I would expect it to give a response, switching out the model for another deployed model works, and manually constructing a request for the openai endpoint https://<foundry-instance>.openai.azure.com/openai/v1/responses works, but then conversations won't work on the relative endpoint.
Additional context
I understand that Astra is new, but I am just as much wondering if this is expected behavior and if we should use the non-project specific openai endpoint.
Describe the bug
GPT-6-astra not working when using the project client.
To Reproduce
Steps to reproduce the behavior:
InternalServerError: Error code: 500 - {'error': {'message': 'The server had an error processing your request. Sorry about that! You can retry your request, or contact us through an Azure support request at: https://go.microsoft.com/fwlink/?linkid=2213926 if you keep seeing this error. (Please include the request ID <redacted> in your email.)', 'type': 'server_error', 'param': None, 'code': 'server_error', 'request_id': '<redacted>'}}Expected behavior
I would expect it to give a response, switching out the model for another deployed model works, and manually constructing a request for the openai endpoint
https://<foundry-instance>.openai.azure.com/openai/v1/responsesworks, but then conversations won't work on the relative endpoint.Additional context
I understand that Astra is new, but I am just as much wondering if this is expected behavior and if we should use the non-project specific openai endpoint.