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

> Source: <https://www.machinebrief.com/news/googles-diffusiongemma-proves-you-dont-need-to-train-from-sc-xvqz>
> Published: 2026-08-09 10:01:26+00:00

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

[The Decoder](https://the-decoder.com)

Instead of [training](/glossary/training) a new model from scratch, Google [DeepMind](/glossary/deepmind) retrofitted Gemma 4 into a [diffusion model](/glossary/diffusion-model) using less than 10 percent of the original training budget. DiffusionGemma generates 256 tokens in parallel instead of one at a time, hitting about 1,500 tokens per second. Quality still trails the original [autoregressive model](/glossary/autoregressive-model) in benchmarks, especially on [reasoning](/glossary/reasoning) tasks.

The article [Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model](https://the-decoder.com/googles-diffusiongemma-proves-you-dont-need-to-train-from-scratch-to-build-a-text-diffusion-model/) appeared first on [The Decoder](https://the-decoder.com).

Get AI news in your inbox

Daily digest of what matters in AI.

## Key Terms Explained

[Autoregressive Model](/glossary/autoregressive-model)

A model that generates output one piece at a time, with each new piece depending on all the previous ones.

[Decoder](/glossary/decoder)

The part of a neural network that generates output from an internal representation.

[DeepMind](/glossary/deepmind)

A leading AI research lab, now part of Google.

[Diffusion Model](/glossary/diffusion-model)

A generative AI model that creates data by learning to reverse a gradual noising process.