Hey, we are a small community team who are trying to create an AI model for our project. We saw that gemini 1.5 is discontinued and went with gemini 2.0 flash-lite. But I’m not sure how to fine tune it as everywhere I go it says it can be fine tuned but also can’t be. And googles resources are conflicting, one one hand, it says it cant be fine-tuned. On the other hand, the python SDK says it can. Anyone have any idea?
1 Like
gemini 2.0 flash-lite can be finetuned using Vertex AI platform.
Here is the link to documentation - Tune Gemini models by using supervised fine-tuning | Generative AI on Vertex AI | Google Cloud
I am from the google team. Let me know if you have tried using Vertex AI platform as @krish_cloudarchitect has suggested.
Alot of users have been able to customize using Vertex AI platform. If you are still unable to figure it out, please paste your entire try with error response and we can try to address the issue.