Error with OpenAI Python Library when Using API with stream=True

API
,
1

Are there any known issues with the OpenAI Python library when using the API with the stream=True parameter?

I’m consistently encountering the following error, while using the library after few hits:

“Must provide an ‘engine’ or ‘deployment_id’ parameter to create a <class ‘openai.api_resources.chat_completion.ChatCompletion’>”.

I would greatly appreciate any suggestions or guidance regarding this. I’m unsure if I’m making any mistakes in my implementation.

2

this is my streaming function in python if it’s of any use to you

def stream_openai_response(prompt):
    response = openai.ChatCompletion.create(
        model='gpt-3.5-turbo',
        messages=[{"role": "user", "content": prompt}],
        max_tokens=1000,
        stream=True,
    )
    for event in response:
        yield event
3

If you can share your function call, that would be really helpful in debugging the problem

4

Thanks @Foxalabs and @udm17 for your suggestions.

this is how my openai call looks like.

def openAI_call(model,message,max_token):
    response = openai.ChatCompletion.create(
        model=model,
        messages=message,
        max_tokens=int(max_token),
        temperature=0.7,
        top_p=1,
        frequency_penalty=0,
        presence_penalty=0,
        n=1,
        stream=True
    )
    for chunk in response:
        choices = chunk.get("choices",[])
        if choices and len(choices)>0 and "delta" in choices[0]:
            delta = choices[0].get("delta",{})
            #print(delta)
            if "content" in delta:
                text  = delta.get("content",{})
                yield str(text + '\n').encode('utf-8')
5

Are you using the async client? if yes, try this -

...
async for chunk in await response:
   choices = chunk.get("choices", [])