Thank you for the follow-up, Paul
Sure, the API key has been tested several times. No impact.

In fact, the key is not even contacter by the plug-in, it seems there is no communication between both. Can this be explained by the fact that the plug-in is looking for the API key in the file "Settings.xml" while it is stored in "Config.xml"?

Can you please check and confirm?

Thank you,

Benoit

Benoit Morelle I will review this with Patrick Andrew Hartnett 

Hi Benoit Morelle ,

Thank you for providing your log files.  I have reviewed them this evening and confirm that you are attempting to access the correct model (e.g. gpt-4).  However, the message that is returned from the sever indicates that you don't have access to the gpt-4 API.

The exact message that is captured from your log file: The model `gpt-4` does not exist or you do not have access to it. Learn more: https://help.openai.com/en/articles/7102672-how-can-i-access-gpt-4

From what you have explained above, you have a paid gpt-4 account and you also confirmed that you recreated the API key and saved it in the OpenAI Translator plugin settings.

Would you be available for a meeting via Teams where you can demo this issue from your sytem?  It will also give me an opportunity to try identify what is the cause..
If yes, let's organize via e-mail.

Please let me know,

Benoit Morelle said:

While the file "Settings.xml" is storing the prompt, not the API key. The API key is stored in the file "Config.xml", instead.

Is it just a detail, or can it explains why the plug-in is not working? Is the plug-in checking/looking for the API in a wrong place, then sending a pop-up error message?

The plugin is persisting the information in different files to separate concerns.  This does not affect the feature functionality. Note: If this was the cause, it would also prevent you from running a query against the model 3.5-turbo

Hi Benoit Morelle , in the meantime, It might also make sense to ask the question to OpenAI support to confirm your account enables you to access gpt-4 via API?

Thank you, Patrick

For a videoconference and to share screen, I have availability:

• this afternoon,
• Monday morning before 11am
• Tuesday, between 10am to 12pm

Is one of these options fit your agenda?

In parallel, I will forward this new question to OpenAI.

Benoit

Hi Benoit Morelle , I've sent a meeting request via e-mail

Hi Benoit Morelle ,

Thank you for your availability this morning to review the issue directly from your environment, allowing us to identify the problem being related to your OpenAI account not enabling you with access to gpt-4 via API.

To check if you can access gpt-4, you can run either of these scripts from your computer:

cURL command

1. Install cURL if not installed on your computer
2. Run the following command from the commandline (cmd).
3. The expected result is a response with the translation suggestion
   

   curl -X POST https://api.openai.com/v1/chat/completions 
   -H "Content-Type: application/json" 
   -H "Authorization: Bearer YOUR_API_KEY" 
   -d "{ \"model\": \"gpt-4\", \"messages\": [{ \"role\": \"user\", \"content\": \"Translate this from en-US to it-IT: The cat is on the table\" }], \"temperature\": 0.7 }"

   
   Note: replace YOUR_API_KEY placeholder with your OpenAI API key.  Replace also the newline characters before running this command from the commandline.

PowerShell script

1. Save the following script in a local file called TestOpenAI.ps1 (e.g. c:\temp\TestOpenAI.ps1)
2. Open a powershell window and run the script c:\temp\TestOpenAI.ps1
3. The expected result is a response with the translation suggestion
   

   # Replace 'YOUR_API_KEY' with the actual API key provided by OpenAI
   $apiKey = 'YOUR_API_KEY'
   
   # Request body with the desired prompt and parameters
   $requestBody = @{
       model = "gpt-4"
       messages = @(
           @{
               role = "user"
               content = "Translate this from en-US to it-IT: The cat is on the table"
           }
       )
       temperature = 0.7
   }
   
   # GPT-4 API endpoint URL for chat completions
   $url = 'https://api.openai.com/v1/chat/completions'
   
   $headers = @{
       'Authorization' = "Bearer $apiKey"
       'Content-Type' = 'application/json'
   }
   
   try {
       $response = Invoke-RestMethod -Uri $url -Headers $headers -Method Post -Body ($requestBody | ConvertTo-Json)
   
       # Display the response or perform actions based on the response
       Write-Host "Response:"
       Write-Host $response
   }
   catch {
       Write-Host "Error occurred: $_.Exception.Message"
   }

   
   Note: replace YOUR_API_KEY placeholder with your OpenAI API key