I'm using OpenAI to learn more about API integration, but I keep running into this code when running the Python program. I asked ChatGPT about the Invalid URL (POST /v1/engines/gpt-3.5-turbo/chat/completions)
error, but it didn't seem to give me the right solutions.
Note: I do have the latest OpenAI package installed (i.e., 0.27.4
).
Code:
import os
import openai
openai.api_key = "sk-xxxxxxxxxxxxxxxxxxxx"
messages = [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Tell me a joke."}
]
response = openai.ChatCompletion.create(
engine="gpt-3.5-turbo",
messages=messages,
max_tokens=50,
n=1,
stop=None,
temperature=0.7,
)
joke = response.choices[0].text.strip()
print(joke)
The ChatGPT API (i.e., the GPT-3.5 API) has a model
parameter (required). The engine
parameter is not a valid parameter for the /v1/chat/completions
API endpoint. See the official OpenAI documentation.
Change this...
engine = "gpt-3.5-turbo"
...to this.
model = "gpt-3.5-turbo"
Also, change this...
joke = response.choices[0].text.strip()
...to this.
joke = response['choices'][0]['message']['content']