Closed Iven2132 closed 3 months ago
The underlying tuning feature does not currently support using images - only text. There's not much we can do about that in the cookbook unfortunately so I'll close this.
I recommend that anyone interested in this feature should file feedback through AI Studio, using Send feedback
under the ⋮
menu.
Looks like it's supported now
https://cloud.google.com/vertex-ai/generative-ai/docs/models/tune_gemini/image_tune
Description of the feature request:
How to fine tune the model with images? fine-tune
What problem are you trying to solve with this feature?
fine-tune the multimodal
Any other information you'd like to share?
No response