google
/

gemma-3-4b-it

Image-Text-to-Text

text-generation-inference

Model card Files Files and versions Community

Finetuning script using HuggingFace

#30

by 2U1 - opened Mar 27

2U1

Mar 27

•

https://github.com/2U1/Gemma3-Finetune

I made a code for who wants to use the huggingface version to finetune, and having difficult using some other frameworks like me.

This code only uses huggingface for fine-tuning the 4B, 12B and 27B.

Also, you can set different learning_rate for vision_tower and language_model. ( Also for the projector)

Feedback and issues are welcome!

Apr 12

Can you provide the notebooks for finetuning for LoRA and QLoRA?

2U1

Apr 15

@Kaith-jeet123 I'll try

Google org 6 days ago

Hi @2U1 ,

Thank you for sharing your fine-tuning code , this is very useful!.

Thanks.

2U1

6 days ago

Thanks for your interest in my work!

I'm currently developing GRPO code for Gemma 3, and I hope it proves to be useful as well.

Google org 6 days ago

@2U1 , That sounds great!. Thank you so much for your continuous interest in the Gemma models. If you required any further assistance please feel free to reach out to us.

Feel free to close this discussion thread if you are satisfied with the above comments.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment