Hi!
I have a license for ChatGpt-4.
I am trying to connect to ChatGPT from the Power Automate flow through the HTTP request, but my authorization does not work. I get an error: “{
“error”: {
“message”: “The model gpt-4 does not exist or you do not have access to it.”,
“type”: “invalid_request_error”,
“param”: null,
“code”: “model_not_found”
}
}”
I was also trying to connect with “gpt-3.5-turbo-instruct” and “gpt-3.5-turbo”, but my authorization was also sanitized.
My example:
Header: Authorization,
Value: Bearer Sk-******
Body:
{
“model”: “gpt-3.5-turbo-instruct”,
“messages”: [
{
“role”: “user”,
“content”: “What is the color of tomato?”
}
],
“max_tokens”: 1000,
“temperature”: 0,
“n”: 1,
“stream”: false,
“stop”: null
}
Could someone please advice how to correct my parameters to connect to ChatGPT?
Your curl request is incorrect. For gpt-3.5-turbo-instruct, the request needs to look as follows (using the completions endpoint https://api.openai.com/v1/completions)
{
"model": "gpt-3.5-turbo-instruct",
"prompt": "What is the color of tomato?",
"max_tokens": 1000,
"temperature": 0,
"stop": null
}
{
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "user",
"content": "What is the color of tomato?"
}
],
"max_tokens": 1000,
"temperature": 0,
"stop": null
}
As for the issue with gpt-4, the curl request would need to look the same as in the case of gpt-3.5.-turbo.
Also note that in order to make a successful API call, you need to make sure that you have pre-funded your OpenAI with at least USD 5 credits and they must still show as active on your usage page.
Hello!
Thank you very much for you very fast response! I really appreciate that.
I made the corrections.
I have also paid for the limits in my account, so I finally got the response from ChatGPT.