I can't believe I'm asking this, but I

I can't believe I'm asking this, but I have no idea now how to call the open ai api any longer

11 comments

Have you tried this?

Plain Text

from llama_index.core.llms import ChatMessage
from llama_index.llms.openai import OpenAI

from llama_index.llms.openai import OpenAI

messages = [
    ChatMessage(
        role="system", content="You are a pirate with a colorful personality"
    ),
    ChatMessage(role="user", content="What is your name"),
]
resp = OpenAI().chat(messages)

print(resp)

TTeemu

https://docs.llamaindex.ai/en/stable/examples/llm/openai.html#openai

ttarpus

will give it a try

ttarpus

why did they change this, I wonder

TTeemu

OpenAI? It should work though

ttarpus

yes, open ai

ttarpus

it's annoying, bro

ttarpus

everyone has to go back and fix their code. like, why

ttarpus

@Teemu so this is what I have reasoned:

in case anyone has to deal with this in the near term:

“import openai”

this has to be commented out.

and replaced with this:

"from openai import OpenAI
client = OpenAI()”

the way we call the openai API has changed from this:

# Call the GPT model
response = openai.ChatCompletion.create(
model='gpt-4-1106-preview',
messages=[{"role": "system", "content": prompt}]
)

to this:

response = client.chat.completions.create(
model='gpt-4-1106-preview',
messages=[{"role": "system", "content": prompt}]
)

the way we process the output has changed from this:

model_output = response['choices'][0]['message']['content']

to this
model_output = response.choices[0].message.content

ttarpus

@Teemu so this is what I have reasoned:

in case anyone has to deal with this in the near term:

1. “import openai”


this has to be commented out. 

and replaced with this:

"from openai import OpenAI
client = OpenAI()”




the way we call the openai API has changed from this:


        # Call the GPT model

response = openai.ChatCompletion.create(
model='gpt-4-1106-preview',
messages=[{"role": "system", "content": prompt}]
)




to this:

response = client.chat.completions.create(
model='gpt-4-1106-preview',
messages=[{"role": "system", "content": prompt}]
)




3. the way we process the output has changed from this:

model_output = response['choices'][0]['message']['content']



to this

model_output = response.choices[0].message.content``

WWhiteFang_Jr

OpenAI updated its library which changed the way OpenAI API calling used to happen. i.e openai.ChatCompletion.create got changed to client.chat.completions.create

As @Teemu mentioned, You can use this if you want to interact with OpenAI

Add a reply

Find answers from the community

I can't believe I'm asking this, but I