I have a use case with video/images + text and generate an answer which Chatgpt does not do very well today. I want to fine-tune the mini model with my own data but found that currently the model does not support video/images fine-tuning. Wondering when OpenAI will support this?
Commenting for thread visibility. It would be helpful to know for project planning purposes.
It is being looking into, but there are no timelines as yet.
2 Likes