Can't get this to work with Azure OpenAI gpt-4o model

[andreasstavrousatalia]

Hello, I'm trying to get this to work but for some reason I get {"error":{"code":"404","message": "Resource not found"}}

We have a gpt-4o deployment on Azure AI Studio

https://azure-openai-[XXXXXX].openai.azure.com/openai/deployments/[DEPLOYMENT_NAME]/chat/completions?api-version=[API_VERSION]

Looking at your azure_provider.py, trying to reverse engineer the way you construct the base url but still I get either resource not found.

ai.Client({
    "azure":{
        "base_url": "https://azure-openai-[XXXXX].openai.azure.com/", # I've tried with /openai/deployments/[DEPLOYMENT_NAME] as well
        "api_key": [API_KEY],
        "api_version": [API_VERSION]
        }}
)

model = "azure:[DEPLOYMENT_NAME]" 

I've verified my credentials are valid, we have been using them in our application for sometime now.

Do I do something wrong in here or is gpt-4o not supported through azure provider? (Line 18-37 azure_provider.py doesn't mention it in the comment)

Thanks!

[Valkea]

I had the same problem, so I forked and solved the problem (the openai models and the other models hosted on Azure doesn't call the same type of URL).

The work is available here : https://github.com/Valkea/aisuite/tree/valkea/azure_openai_fix