Hugging Face (Twitter)

RT @osanseviero: Introducing T5Gemma 2, the next generation of encoder-decoder models 🚀

Built on top of Gemma 3, we were able to build compact models at sizes of 270m-270m, 1B-1B, and 4B-4B sizes.

While most models today are decoder-only, T5Gemma 2 is the first (I'm aware of) multimodal, long-context, and heavily multilingual (140 languages) encoder-decoder model out there.

We hope this model enables the model research community as well as the community of devs ready to explore with new architectures.

Blog: https://blog.google/technology/developers/t5gemma-2
Models: https://huggingface.co/collections/google/t5gemma-2
Paper: arxiv.org/abs/2512.14856