TypeError: 'LegacyAPIResponse' object is not iterable

[psymbio]

Confirm this is an issue with the Python library and not an underlying OpenAI API

• This is an issue with the Python library

Describe the bug

Not able to get a streaming response because 'LegacyAPIResponse' object is not iterable.

To Reproduce

The following code:

import os
import openai
from openai import AzureOpenAI

os.environ['AZURE_OPENAI_API_KEY'] = "xxx"
client = AzureOpenAI(
    api_version="2023-07-01-preview",
    azure_endpoint="https://xxx.openai.azure.com/",
)

stream = client.chat.completions.with_raw_response.create(
    messages=[{
        "role": "user",
        "content": "sing me a song",
    }],
    model="gpt-35-turbo",
    max_tokens=30,
    temperature=0.7,
    stream=True
)
print(stream)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

Results in:

<APIResponse [200 OK] type=<class 'openai.types.chat.chat_completion.ChatCompletion'>>
Traceback (most recent call last):
  File "test.py", line 27, in <module>
    for chunk in stream:
TypeError: 'LegacyAPIResponse' object is not iterable

Code snippets

No response

OS

WSL2 Ubuntu

Python version

Python v3.9.7

Library version

openai-python v1.10.0

[RobertCraigie]

This is because you need to actually call .parse() to get the stream class, e.g.

response = client.chat.completions.with_raw_response.create(
    messages=[{
        "role": "user",
        "content": "sing me a song",
    }],
    model="gpt-35-turbo",
    max_tokens=30,
    temperature=0.7,
    stream=True
)
stream = response.parse()
for chunk in stream:
  ...

[psymbio]

Working code:

import os
import openai
from openai import AzureOpenAI
assert openai.__version__ == "1.10.0"

os.environ['AZURE_OPENAI_API_KEY'] = "xxx"

client = AzureOpenAI(
    api_version="2023-07-01-preview",
    azure_endpoint="https://xxx.openai.azure.com/",
)

response = client.chat.completions.with_raw_response.create(
    messages=[{
        "role": "user",
        "content": "sing me a song",
    }],
    model="gpt-35-turbo",
    max_tokens=30,
    temperature=0.7,
    stream=True
)
stream = response.parse()
for chunk in stream:
    if (len(chunk.choices)) > 0:
        print(chunk.choices[0].delta.content or "", end="")