OperationNotSupported when trying to fun completion with gpt-4o

Question

I've created a gpt-4o deployment and now trying to run a test program generated here:

https://learn.microsoft.com/en-us/azure/ai-services/openai/quickstart?pivots=programming-language-python&tabs=command-line%2Cpython-new

import os
from openai import AzureOpenAI
    
client = AzureOpenAI(
    api_key=os.getenv("AZURE_OPENAI_API_KEY"),  
    api_version="2024-02-01",
    azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT")
    )
    
deployment_name='REPLACE_WITH_YOUR_DEPLOYMENT_NAME' #This will correspond to the custom name you chose for your deployment when you deployed a model. Use a gpt-35-turbo-instruct deployment. 
    
# Send a completion call to generate an answer
print('Sending a test completion job')
start_phrase = 'Write a tagline for an ice cream shop. '
response = client.completions.create(model=deployment_name, prompt=start_phrase, max_tokens=10)
print(start_phrase+response.choices[0].text)

However I'm getting an error:

openai.BadRequestError: Error code: 400 - {'error': {'code': 'OperationNotSupported', 'message': 'The completion operation does not work with the specified model, gpt-4o. Please choose different model and try again. You can learn more about which models can be used with each operation here: https://go.microsoft.com/fwlink/?linkid=2197993.'}}

Answer 1

Hello Romanowski, Tomasz,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand that you are having issue with running a completion operation on a GPT-4o model deployed in Azure OpenAI Service and this makes you to request for an assistance in identifying a suitable model and modifying your code to work with the model.

Solution

The error message you're seeing indicates that the gpt-4o model does not support the completion operation you're trying to perform. This is a known limitation of the gpt-4o model in the Azure OpenAI service and you can read more from this Azure OpenAI documentation: https://go.microsoft.com/fwlink/?linkid=2197993

To resolve this issue, you need to ensure that you are using a model that supports the completion operation. According to the Azure OpenAI documentation, certain models like gpt-35-turbo-instruct are recommended for such tasks. This is how you can modify your code to use the gpt-35-turbo model:

import os
from openai import AzureOpenAI
# Set environment variables
os.environ["AZURE_OPENAI_API_KEY"] = "your_azure_openai_api_key"
os.environ["AZURE_OPENAI_ENDPOINT"] = "your_azure_openai_endpoint"
# Initialize the client
client = AzureOpenAI(
    api_key=os.getenv("AZURE_OPENAI_API_KEY"),
    api_version="2024-02-01",
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT")
)
# Use the correct deployment name for a supported model
deployment_name = 'gpt-35-turbo-instruct-deployment'  # Replace with your deployment name
# Send a completion call to generate an answer
print('Sending a test completion job')
start_phrase = 'Write a tagline for an ice cream shop. '
response = client.completions.create(model=deployment_name, prompt=start_phrase, max_tokens=10)
print(start_phrase + response.choices[0].text)

I replaced gpt-35-turbo with your actual deployment name if you have deployed the gpt-35-turbo model. If you haven't, you might need to deploy it first.

References

Source: This is the current limitation. Assistants doesn't support GPT-4o yet. Accessed, 6/8/2024.

Source: Work with the GPT-35-Turbo and GPT-4 models - Azure OpenAI Service. Accessed, 6/8/2024.

Source: [Azure OpenAI documentation.] Accessed, 6/8/2024.

Accept Answer

I hope this is helpful! Do not hesitate to let me know if you have any other questions.

** Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful ** so that others in the community facing similar issues can easily find the solution.

Best Regards,

Sina Salam

Answer 2

This worked fine for me.

response = client.chat.completions.create(
model=deployment_name,

messages=[

    {"role": "system", "content": "YOU ARE USEFUL ASSISTANT"},

    {"role": "user", "content": "How are you"}

]

)

print(response.choices[0].message.content)