[Issue]: zero cost in chat_res.cost while already called the openai gpt-3.5-turbo

[yanan1116]

Describe the issue

hi, i am trying to follow examples on my own workstation.
such as this one : https://github.com/microsoft/autogen/blob/main/notebook/agentchat_auto_feedback_from_code_execution.ipynb

the issue is that there is zero cost in Cost info: Cost info: ({'total_cost': 0, 'gpt-3.5-turbo-0125': {'cost': 0, 'prompt_tokens': 3941, 'completion_tokens': 587, 'total_tokens': 4528}}, {'total_cost': 0, 'gpt-3.5-turbo-0125': {'cost': 0, 'prompt_tokens': 3941, 'completion_tokens': 587, 'total_tokens': 4528}})

actually, i use gpt-3.5-turbo-0125 in config_list instead of gpt-4; while other codes are the same with the example code

execution result:

user_proxy (to assistant):

What date is today? Compare the year-to-date gain for META and TESLA.

---

assistant (to user_proxy):

To get today's date, we can use Python. Let's first get the current date and then we can proceed to compare the year-to-date gain for META and TESLA.

Here is the Python code to get today's date:

# filename: get_date.py
from datetime import date

today = date.today()
print(today)

Please run the code above to get today's date. After that, we can proceed with comparing the year-to-date gain for META and TESLA.

---

EXECUTING CODE BLOCK 0 (inferred language is python)...
user_proxy (to assistant):

exitcode: 0 (execution succeeded)
Code output:
2024-03-13

---

assistant (to user_proxy):

Great! Today's date is March 13, 2024.

Now, to compare the year-to-date gain for META and TESLA, we would typically need the stock prices at the beginning of the year and the current stock prices for both companies. Since I don't have real-time data, I will provide a general approach using placeholder values for demonstration purposes.

Let's assume the following placeholder values:

• META stock price at the beginning of the year: $100
• META current stock price: $120
• TESLA stock price at the beginning of the year: $800
• TESLA current stock price: $900

We will calculate the year-to-date gain for both META and TESLA based on these placeholder values.

Here is the Python code to compare the year-to-date gain for META and TESLA:

# Placeholder values for demonstration
meta_start_price = 100
meta_current_price = 120
tesla_start_price = 800
tesla_current_price = 900

# Calculate year-to-date gain for META and TESLA
meta_ytd_gain = ((meta_current_price - meta_start_price) / meta_start_price) * 100
tesla_ytd_gain = ((tesla_current_price - tesla_start_price) / tesla_start_price) * 100

print("META year-to-date gain: {:.2f}%".format(meta_ytd_gain))
print("TESLA year-to-date gain: {:.2f}%".format(tesla_ytd_gain))

Please run the code above to compare the year-to-date gain for META and TESLA based on the placeholder values. If you have the actual stock prices, you can replace the placeholder values with the real data for a more accurate comparison.

---

EXECUTING CODE BLOCK 0 (inferred language is python)...
user_proxy (to assistant):

exitcode: 0 (execution succeeded)
Code output:
META year-to-date gain: 20.00%
TESLA year-to-date gain: 12.50%

---

assistant (to user_proxy):

Based on the placeholder values provided, the year-to-date gain for META is 20.00% and for TESLA is 12.50%.

If you have the actual stock prices, you can replace the placeholder values in the code with the real data to get a more accurate comparison.

Is there anything else you would like assistance with? If not, please type "TERMINATE".

---

user_proxy (to assistant):

---

assistant (to user_proxy):

TERMINATE

---

Steps to reproduce

No response

Screenshots and logs

Additional Information

autogen version: 0.2.17

[yiranwu0]

This is because "gpt-3.5-turbo-0125" is not stored in OAI_PRICE, but it is added now in #2332