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, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("google/gemma-4-12B-it") model = AutoModelForImageTextToText.from_pretrained("google/gemma-4-12B-it") - Notebooks
- Google Colab
- Kaggle
Good, Google
Happy to see a 12b dense model. Hope you have another bigger MoE unreleased :)
Most users like a 9b ~ 12b dense model because it can be easily fine tuned on a single node!
Happy to see a 12b dense model. Hope you have another bigger MoE unreleased :)
The MOE table is looking real empty rn with just 1 model wink wink
I will name my next child Jeffdean if we get a 124B MoE
Hi all,
Thanks for the feedback! Let us know if you run into any issues while setting it up, and feel free to share if you end up building something cool with it.
Hi all,
Thanks for the feedback! Let us know if you run into any issues while setting it up, and feel free to share if you end up building something cool with it.
Gemma 4:124b. Please and Thank You.