-
Notifications
You must be signed in to change notification settings - Fork 140
Closed
Labels
good first issueGood for newcomersGood for newcomerskind/bugSomething isn't workingSomething isn't working
Description
When using cagent
with azure
provider, the models deployed with "dot" in the name aren't supported yet:
model config
models:
azure_model:
provider: azure
model: gpt-4.1
# TODO: Replace with your Azure OpenAI base URL
base_url: https://your-project-resource.cognitiveservices.azure.com
provider_opts:
api_version: 2025-01-01-preview
when running with above model configuration I get following error:
time=2025-09-16T15:26:50.819+02:00 level=ERROR msg="OpenAI stream creation failed" error="error, status code: 404, status: 404 Not Found, message: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again." model=gpt-4.1
time=2025-09-16T15:26:50.819+02:00 level=ERROR msg="Failed to create chat completion stream" agent=root error="error, status code: 404, status: 404 Not Found, message: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again."
Using model name gpt-4-1
fixes the issue!
The issue seem to be coming from sashabaranov/go-openai#978 but makes sense have an issue in cagent
users for tracking!
Metadata
Metadata
Assignees
Labels
good first issueGood for newcomersGood for newcomerskind/bugSomething isn't workingSomething isn't working