[Python] Best way to stream using langchain `llm.with_structured_output`

import asyncio
from langchain_core.pydantic_v1 import BaseModel, Field
from langchain_openai import ChatOpenAI

class Joke(BaseModel):
setup: str = Field(description="The setup of the joke")
punchline: str = Field(description="The punchline to the joke")

# Define the model with structured output
model = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0, streaming=True)
structured_llm = model.with_structured_output(Joke)

def stream_joke():
# Invoke the model and stream the response
response = structured_llm.stream("Tell me a joke about cats")

# Initialize an empty Joke object
joke = Joke(setup="", punchline="")

# Stream the response
for part in response:
if 'setup' in part:
joke.setup += part['setup']
print(f"Setup: {joke.setup}")
if 'punchline' in part:
joke.punchline += part['punchline']
print(f"Punchline: {joke.punchline}")

# Run the streaming joke function
stream_joke()


What is the best practice to stream using langchain llm.with_structured_output? This just doesn't work.

Continue reading...