Hi everyone!

I am developing a RAG chatbot. I have tried different models using the AzureOpenAI center. Therefore, I had to change to a different region and therefore had to set up a new Azure OpenAI account than that I was using initially.

With the code for the chatbot, I have already managed to get through the error I am getting right now:

I am always getting the same error:

NotFoundError: Error code: 404 - {'error': {'code': '404', 'message': 'Resource not found'}}

With the new and the old Azure OpenAI account, I am not able to resolve this error. I have checked the key and endpoint are correct etc. I am still getting this error.

I also tried minimal code example like the one below.

I'd appreciate any help!

Thanks a lot in advance!

Best,

Johanna

Minimal code example(https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/switching-endpoints?source=recommendations):

Load environment variables for Azure OpenAI

from langchain.chat_models import AzureChatOpenAI

from langchain.schema import HumanMessage

os.environ['AZURE_OPENAI_API_KEY'] = 'xxxxxxxxxxx'

os.environ['AZURE_OPENAI_ENDPOINT'] = 'https://****.openai.azure.com/'

import os

from openai import AzureOpenAI

client = AzureOpenAI(

api_key=os.getenv("AZURE_OPENAI_API_KEY"),  

api_version="2024-05-13", #used a different version number for the other Azure OpenAI account 

azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT")

)

completion = client.completions.create(

model="myexamplename", # This must match the custom deployment name you chose for your model.

prompt="<prompt>"

)

chat_completion = client.chat.completions.create(

model="gpt-4o", # model = "deployment_name".

messages="<messages>"

)

embedding = client.embeddings.create(

model="text-embedding-ada-002", # model = "deployment_name".

input="<input>"

)

gives:

NotFoundError: Error code: 404 - {'error': {'code': '404', 'message': 'Res

Hi @Kemper Johanna

Greetings & Welcome to Microsoft Q&A forum! Thanks for posting your query!

Even though you have verified the key and endpoint, the "NotFoundError" suggests that the Azure OpenAI service is unable to locate the requested resource.

Here are some troubleshooting steps that might help you in resolving the query:

1. Match deployment names: Make sure the deployment_name in your code exactly matches the "Model deployment name" in the Azure portal. Remember, this is case-sensitive.
2. Review Authorization: Ensure that your application has the necessary permissions to access the deployment. This might involve checking the role-based access control (RBAC) settings in the Azure portal.
3. Check API version compatibility: Ensure that the openai_api_version you are using is supported by both the openai package and the Azure OpenAI service.

Make sure you are using the right API key, endpoint, and API version in your chatbot code. It's important to double-check these details against what's in your Azure OpenAI account. Also, check if you've deployed the specific GPT model you need for your resource. If it's not deployed yet, you'll have to do that before you can start using it in your chatbot.

For additional information please refer the following links:

• https://github.com/langchain-ai/langchain/issues/14018
• https://github.com/langchain-ai/langchain/issues/13284
• https://community.openai.com/t/azure-open-ai-api-endpoint-404-resource-not-found/650881

I hope this information helps, please do let us know if you have any further queries.

Thank you.

Hi @Kemper Johanna

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.

Hello @Kemper Johanna, as hinted by Chandra Boorla very likely it's a mismatch of the Azure OpenAI deployment's name. In the "model=" assignment, will you please enter the custom name of your deployment, not the type of Azure OpenAI model.

Hi @Kemper Johanna

Just checking in to see on the last response. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.

Thank you.

Hi Kemper

Try out my code below, it worked for me.

the issue is from api version most likely. You had api_version as '2024-05-13' and i get same error as your 404. if updated to 2024-07-01-preview, the call will get a response.

You might be using the samples from here.

https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/switching-endpoints

import os
from openai import AzureOpenAI

client = AzureOpenAI(
    api_key='53f4ec2340964xxxxx',
    api_version="2024-07-01-preview",  # you had '
    azure_endpoint='https://xxxxx.openai.azure.com/',
)


response = client.chat.completions.create(
                model="completions", # your deployment model name
                messages=[
                    {"role": "system", "content": "You are a helpful assistant."},
                    {"role": "user", "content": "Who won the world series in 2020?"}
                ]
            )

print(response)