openai-cookbook icon indicating copy to clipboard operation
openai-cookbook copied to clipboard

[PROBLEM] GPT 4o mini unable to process images

Open Kp101coder opened this issue 1 year ago • 4 comments
trafficstars

Identify the file to be fixed openai-cookbook/blob/main/examples/gpt4o/introduction_to_gpt4o.ipynb

Describe the problem In the cookbook, it is stated that the new 4o mini that the model is able to accept inputs of text and image. However when using the code provided in openai-cookbook/blob/main/examples/gpt4o/introduction_to_gpt4o.ipynb, section Base64 Image Processing, the AI simply returns responses similar to "I am unable to view the image...".

Additional context By changing the model to GPT 4o and nothing else (changing the model="gpt-4o-mini" to model = gpt-4o"), the desired response is achieved. Either there is an issue with GPT 4o mini API, or the cookbook has mistakenly labeled it as being able to support image inputs.

Kp101coder avatar Jul 19 '24 06:07 Kp101coder