API Usage shows GPT-4-0613 instead of GPT-4

API
1

I’ve been using GPT-4 since April and it’s been working fine, but when I check my API usage (Daily usage breakdown, OpenAI Platform), it says I am requesting the GPT-4-0613 model instead of the GPT-4 model.

In my completion request I have always been specifying the model GPT-4, but now I wonder if my responses are from the GPT-4 or GPT-4-0613 model.

Anyone else noticed this, or what am I missing here?

My python chat completion request code below:
def gpt4_chat_completion(
messages, #previously “prompt”
model=“gpt-4”,
# model=‘gpt-3.5-turbo’,
temp=0,
top_p=1.0,
tokens=4096,
freq_pen=0.0,
pres_pen=0.0,
stop=[‘<>’]):

# Ensure the messages are ASCII-encoded
for i, message in enumerate(messages):
    messages[i]['content'] = message['content'].encode(encoding='ASCII', errors='ignore').decode()

# prompt = prompt.encode(encoding='ASCII',errors='ignore').decode()

# Call the OpenAI API for text completion
completion = openai.ChatCompletion.create(
    model=model,
    messages=messages,
    temperature=temp,
    max_tokens=tokens,
    top_p=top_p,
    frequency_penalty=freq_pen,
    presence_penalty=pres_pen,
    stop=stop)

# Extract the generated text from the completion response
text = completion['choices'][0]['message']['content'].strip()
return text, model, temp, tokens

When I print my model lists, I get a long list, and the entries for GPT-4 and GPT-4-0613 look like this when I run the following in python:
model_list= openai.Model.list()
print(model_list)

{
    "id": "gpt-4-0613",
    "object": "model",
    "created": 1686588896,
    "owned_by": "openai",
    "permission": [
        {
            "id": "modelperm-rTqw0DHQTpqP0Y9y0I3wsgsx",
            "object": "model_permission",
            "created": 1695226816,
            "allow_create_engine": false,
            "allow_sampling": false,
            "allow_logprobs": false,
            "allow_search_indices": false,
            "allow_view": false,
            "allow_fine_tuning": false,
            "organization": "*",
            "group": null,
            "is_blocking": false
        }
    ],
    "root": "gpt-4-0613",
    "parent": null
},
{
    "id": "gpt-4",
    "object": "model",
    "created": 1687882411,
    "owned_by": "openai",
    "permission": [
        {
            "id": "modelperm-6FFWHA4eLhaNkkCaNQYiVryW",
            "object": "model_permission",
            "created": 1695226861,
            "allow_create_engine": false,
            "allow_sampling": false,
            "allow_logprobs": false,
            "allow_search_indices": false,
            "allow_view": false,
            "allow_fine_tuning": false,
            "organization": "*",
            "group": null,
            "is_blocking": false
        }
    ],
    "root": "gpt-4",
    "parent": null
},
2

They are the same thing.

June:

Model deprecations

Today, we’ll begin the upgrade and deprecation process for the initial versions of gpt-4 and gpt-3.5-turbo that we announced in March. Applications using the stable model names (gpt-3.5-turbo, gpt-4, and gpt-4-32k) will automatically be upgraded to the new models listed above on June 27th. For comparing model performance between versions, our Evals library supports public and private evals to show how model changes will impact your use cases.

1 Like