I encountered the following error when trying to use LiteLLM. [ADDED Log File]

[durul]

Hello,

Here is my error message.

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

Additional details:

1. LiteLLM version: 1.48.1
2. Python version: 3.12.4
3. Python Path: /opt/homebrew/anaconda3/bin/python
4. Operating System: macOS 15.0 arm64
5. Steps to reproduce:

• I uploaded a PDF file.
• I entered my OpenAI API key.
• I selected the model "o1-preview-2024-09-12".
• I left all the other features as a default.

Here is the log file
litellm_debug.log

I've already tried enabling verbose logging with litellm.set_verbose=True, but I'm unsure how to proceed. Any guidance on resolving this issue would be greatly appreciated.

[aymiee]

I have OpenAI api key, but I am not a Tier 5. I thought I read that API usage for o1-preview and 01-mini were for Tier 5. https://platform.openai.com/docs/guides/reasoning. I changed the model to GPT-4-Turbo and no more issues. Is this perhaps your issue, too?

[durul]

Hello, I am Tier 1. Can it be related to my tier?

[aymiee]

I believe so. As of 9/12, Tier 1 and does not have access to o1-preview and 01-mini via the API.

[durul]

I believe so. As of 9/12, Tier 1 and does not have access to o1-preview and 01-mini via the API.

🥲 Is there any way would you recommend? I could also try it in my local, but nothing changed.

[moebiussurfing]

I've already tried enabling verbose logging with litellm.set_verbose=True, but I'm unsure how to proceed. Any guidance on resolving this issue would be greatly appreciated.

Sorry @durul , what file and where do you have edited this flag?

[durul]

I've already tried enabling verbose logging with litellm.set_verbose=True, but I'm unsure how to proceed. Any guidance on resolving this issue would be greatly appreciated.

Sorry @durul , what file and where do you have edited this flag?

You can find the updated file below with literally.set_verbose=True.

app.py.zip

[durul]

I have also attached a log file. Maybe it can help turn on the lights for someone to find the solution.

[moebiussurfing]

thanks @durul

it seems on the log that you are selecting another model instead of openai/custom_model.
The model o1-preview-2024-09-12 does not exist or you do not have access to it.'

Are you trying to use your payed openAI account or a local alternative?
Maybe not all models are available for some user tiers or maybe not implemented...

In the case you are trying as local model:
Did you checked this issue?
#1

[durul]

thanks @durul

it seems on the log that you are selecting another model instead of openai/custom_model. The model o1-preview-2024-09-12 does not exist or you do not have access to it.'

Are you trying to use your payed openAI account or a local alternative? Maybe not all models are available for some user tiers or maybe not implemented...

In the case you are trying as local model: Did you checked this issue? #1

I tried the current and all previous models. but the result is nothing changed.