Gemma 4 Soft Tokens: The Rise and Fall of 16x16 Words ⚡👀 (opens in new tab)

This is a submission for the The road to vision capabilities in the Gemma family has been an interesting one. The first and second generations of Gemma models did not include native vision support. Instead, multimodal capabilities were introduced through the PaliGemma models. It wasn’t until Gemma 3 that we began to see native vision capabilities integrated directly into the Gemma. Even then, those capabilities were reserved for the larger variants. With Gemma 4, that changes. Every variant i...