Assistant API response missing code

API
1

Hi all,
I created an assistant from the browser and it works really well. I wrote some python code following the documentation to send a message the assistant and get a response back. Everything works well, but the assistant in the browser responds back with code (as it should), and the same assistant from the API will omit all of that out. feels like it’s some type of rate limit response? I’m not sure why but the conversation with the API assistant is only theory based.

Python code:
import json
from openai import OpenAI
import time
import os

#store somewhere more secure, remove from scipt
client = OpenAI(
api_key=‘sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx’,
)

assistant1_id = ‘xxxxxxxxxxxxxxxxxxxxxxxxx’

def create_thread(user_message):
thread = client.beta.threads.create()

try:
    # Add user message to the thread
    message = client.beta.threads.messages.create(
        thread_id=thread.id,
        role="user",
        content=user_message
    )
    print("### message inside send_message is:", message)
    create_run(thread.id)

except Exception as e:
    print(f"An error occurred while sending the message: {e}")
    return None

def create_run(thread_ID):

try:
    # Run assistant
    run = client.beta.threads.runs.create(
        thread_id=thread_ID,
        assistant_id=assistant1_id
    )
    print("### run from create_run is:", run)
    display_assistant_response(thread_ID,run.id)
    
except Exception as e:
    print(f"An error occurred while sending the message: {e}")
    return None

def display_assistant_response(thread_ID, run_id):
try:
# Display assistant response
run = client.beta.threads.runs.retrieve(
thread_id=thread_ID,
run_id=run_id
)
#print(“### run from display_assistant_response is:”, run)

    count = 0
    while(run.status == "queued" or run.status == "in_progress" and count < 5):
        time.sleep(1)
        run = client.beta.threads.runs.retrieve(
            thread_id=thread_ID,
            run_id=run_id
        )
        count = count + 1        
    
    # Retrieve messages from thread after run complete
    messages = client.beta.threads.messages.list(
        thread_id=thread_ID
    )
    
    print(f"---------------------------------------------")
    for message in messages.data:
        for content in message.content:
            if content.type == 'text' and content.text and content.text.value:
                value = content.text.value
                print("### Extracted value:", value)

                # Add the extracted value to the conversation history
                if value:  # Additional check to ensure value is not null
                    system_message = {"role": "system", "content": value}
                else:
                    print("Warning: Null value encountered in message content.")
    
    display_gpt_response(value)
    
except Exception as e:
    print(f"An error occurred while sending the message: {e}")
    return None

def display_gpt_response(value):
try:
response = client.chat.completions.create(
model=“gpt-4-1106-preview”,
messages=[
{“role”: “system”, “content”: value}
]
)
#print(“### response 1 from gpt is:”, response.choices[0].message.content)
print(“### response 1 from gpt is:”, response)

except Exception as e:
    print(f"An error occurred while sending the message: {e}")
    return None

def main():
print(“hello!”)
user_message= “create a small LAN network.”

create_thread(user_message)     

print("end")

if name == “main”:
main()

2

I think you are giving only 5 seconds for the assistant and retrieving messages and printing it. Your thread might need more than 5 seconds to add a message to thread messages. Remove count < 5 and it should work fine I guess.

1 Like
3

I see, that was indeed the problem. Thank you!