drifting ice
it used to work with davinci 003, but as soon as i replaced the model with gpt-3.5-turbo, it has stopped working completely and gives me this error. i have been trying for a while to fix it but i'm not good with python and the error doesn't really make sense to me
openai.error.InvalidRequestError: This is a chat model and not supported in the v1/completions endpoint. Did you mean to use v1/chat/completions?
naive sequoia
Try https://api.openai.com/v1/chat/completions as an endpoint
tame cove
Ya you can't just flip models. You need to update openai, then you need to check the docs and change the format. It doesn't just accept a prompt now, it needs you to send it an array.
oblique sedge
This works for me:
const fetch = require("node-fetch");
const OPENAI_API_KEY = YOUR_API_KEY;
async function chatgptCompletion() {
const requestOptions = {
method: "POST",
headers: {
Authorization: Bearer ${OPENAI_API_KEY},
"Content-Type": "application/json",
},
body: JSON.stringify({
model: "gpt-3.5-turbo",
messages: [{ role: "user", content: "What is the OpenAI mission?" }],
}),
};
fetch("https://api.openai.com/v1/chat/completions", requestOptions)
.then((response) => response.json())
.then((result) => console.log(result.choices[0]))
.catch((error) => console.log("error", error));
}
chatgptCompletion();
silver lynx
^^ Yup, you can't just flip models
vernal jacinth
in python, after defined openai.api_endpoint = "https://api.openai.com/v1/chat/completions", and using "gpt-3.5-turbo", still keep getting "openai.error.InvalidRequestError: This is a chat model and not supported in the v1/completions endpoint. Did you mean to use v1/chat/completions?"
import openai
import os
import openai
import os
openai.api_key = os.environ["OPENAI_API_KEY"]
openai.api_endpoint = os.environ.get("OPENAI_API_ENDPOINT", "https://api.openai.com")
#model_engine = "text-codex-002"
model_engine = "code-davinci-edit-001"
model_engine = "babbage-code-search-text"
model_engine = "gpt-3.5-turbo"
openai.api_endpoint = "https://api.openai.com/v1/chat/completions"
prompt = 'generate go code:\npackage main\n\nimport "fmt"\n\nfunc main() {\n\tfmt.Println("Hello, world!")\n}\n\n'
response = openai.Completion.create(
engine=model_engine,
prompt=prompt,
max_tokens=1024,
n=1,
stop=None,
temperature=0.7,
)
choices = response.choices[0].text.strip().split('\n')
print(choices[0])
anything wrong here?
i am trying those models fetched from model list, none of them worked so far
tribal beacon
Use the openai.ChatCompletion function instead.
Also instead of choices[0].text it'll be choices[0].content
@vernal jacinth
@drifting ice your problem is likely the same as James'. There's a difference between "Completion" and "ChatCompletion"
vernal jacinth
Traceback (most recent call last):
File "/Users/jamesqu/mydev/openai/python/codex.go.hello.py", line 15, in <module>
response = openai.ChatCompletion.create(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/openai/api_resources/chat_completion.py", line 25, in create
return super().create(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/openai/api_resources/abstract/engine_api_resource.py", line 153, in create
response, _, api_key = requestor.request(
^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/openai/api_requestor.py", line 226, in request
resp, got_stream = self._interpret_response(result, stream)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/openai/api_requestor.py", line 620, in _interpret_response
self._interpret_response_line(
File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/openai/api_requestor.py", line 683, in _interpret_response_line
raise self.handle_error_response(
openai.error.InvalidRequestError: Invalid URL (POST /v1/engines/code-davinci-edit-001/chat/completions)
this time, got Invalid URL error for those models tried so far
think one step forward