Can not access GPT-4o - Issue with billing?

API
1

I trying to access gpt-4o from Python code, and using the sample code below to test it:

from openai import OpenAI

MODEL = "gpt-4o"


client = OpenAI(api_key="XXXXXXXXXXXX")

# Basic Chat

completion = client.chat.completions.create(
    model = MODEL,
    messages=[
        {"role":"system", "content":"You are a helpful assistant. Help me with my math homework"},
        {"role":"user", "content":"Hello! Can you solve 9+11?"}
    ]
)

print(f"Assistant: {completion.choices[0].message.content}")

At the end I am getting

Error code: 404 - {'error': {'message': 'The model `gpt-4o` does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

My question is, although it is announced it is going to be free to access GPT-4o, does it cover all sorts of access or only via the browser on the chat window?

2

I tested your code and it works fine on my end. Please check the permissions of your API key to ensure it has access to the GPT-4o model.

image
image968×663 28.4 KB

3

Thanks @tiansenxu , just to confirm, you are also on free tier (never made any credit top up or so) ?

5

You’re welcome. I am using a paid account. Regarding your question, the free access might be for specific users, such as new registrants. Although I’m not entirely sure, I recall that new users often receive some free credits or access privileges.。

6

I have exactly the same problem - have enough credits and code was working before with v3.5
+updated to the latest python openai package version, didn’t solve the problem

7

I am using paid account and generated my api_key just today.
When I checked my project api keys permissions I can see “All”

openai.NotFoundError: Error code: 404 - {‘error’: {‘message’: ‘The model gpt-4o does not exist or you do not have access to it.’, ‘type’: ‘invalid_request_error’, ‘param’: None, ‘code’: ‘model_not_found’}}

python code is just similar:

from openai import OpenAI

model=“gpt-4o”
client = OpenAI(api_key=“sk-proj-gC6fUxxxxxxxxxxx”)

completion = client.chat.completions.create(
model=model,
messages=[
{“role”:“system”,
“content”: “You are a helpful assitance. He me to solve my math homework!”},

    { "role":"user",
     "content": "Hello! could you solve 2+2 for me?"}
]

)

print("Assitant: "+ completion.choices[0].message.content)

8

You might need to wait until you hit billing Tier 1 or Tier 2… I’d check the docs.

1 Like
9

Thanks @PaulBellow
After checking the tiers and limits page, it seems like the case for me.

10

@PaulBellow where can I check my limit … what should to do hit billing tier 1 or tier2? which docs are you referring . apart from being gptplus user should I need toapy additionally to access apis? thx

11

https://platform.openai.com/settings/organization/limits

13

I hope you have created a project and then, API key.
Follow these steps :

  1. Login to your OpenAI account
  2. go to settings option (available on upper right corner)
  3. under project tab( available in the left pane), go to Limits option
  4. after clicking Limits, you’ll find the Allowed models in Model usage. just click edit button, and add the model you want to use through API.