# Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model

> Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model Instead of training a new model from scratch, Google DeepMind retrofitted Gemma 4 into a diffusion model.

- **Source:** [The-decoder](https://the-decoder.com/googles-diffusiongemma-proves-you-dont-need-to-train-from-scratch-to-build-a-text-diffusion-model/?utm_source=gearopen.com&utm_medium=referral&utm_campaign=feed&utm_content=6a787022658b89ae795dc01c)
- **Published:** 2026-08-09
- **Category:** AI & Bots
- **Tags:** #google #nvidia
- **Canonical:** http://localhost:4321/a/6a787022658b89ae795dc01c

---

_GearOpen's distilled summary. The full original article is published at The-decoder (source link above); GearOpen does not republish the source's full text._