T5Gemma 2

The second generation of open encoder-decoder Gemma models built on Gemma 3, in 270M-270M, 1B-1B and 4B-4B sizes. The first multimodal, long-context encoder-decoders in the family: they read images alongside text, hold up to 128K tokens of context and cover over 140 languages.

Открытые веса, в Vertex AI Model Garden с 18.12.2025.

Understands

text, images

Produces

text, code

Contents verified with the vendor: 2026-08-18

Included in subscriptions

No catalogue entry marks this model as part of a plan yet. It will appear here once the plan contents are confirmed with the vendor.