Instructions to use google/gemma-4-12B-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use google/gemma-4-12B-it with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("google/gemma-4-12B-it") model = AutoModelForMultimodalLM.from_pretrained("google/gemma-4-12B-it") - Notebooks
- Google Colab
- Kaggle
Gemma 4 124b
needs no further explanation
I hold the same opinion with you.
we wait the 124B weight open here and forever
Hi @FusionCow
We appreciate the support! Let us know if you run into any issues or have a chance to build something cool with google/gemma-4-12B-it.
Hi @FusionCow
We appreciate the support! Let us know if you run into any issues or have a chance to build something cool with google/gemma-4-12B-it.
Device models are cool but something large enough to serve users would certainly entice.
Specifically, Gemma 4:124b, that would be very enticing.
Noob question as I am new to this - what is difference between gemma-4-12b and gemma-4-12b-it ? Somehow cant find any difference in description. Thanks
Noob question as I am new to this - what is difference between gemma-4-12b and gemma-4-12b-it ? Somehow cant find any difference in description. Thanks
Here it = instruction tuned meaning it has been run through an additional training process so it understands the chat template (i.e. can have a conversation). The non-it model hasn't been run through this and will just auto-complete text.
lol, you guys want to open source Gemini now