The Decoder Instead of training a new model from scratch, Google DeepMind retrofitted Gemma 4 into a 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 in benchmarks, especially on reasoning tasks.
The article Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model appeared first on The Decoder.
Get AI news in your inbox
Daily digest of what matters in AI.
Key Terms Explained #
Autoregressive Model A model that generates output one piece at a time, with each new piece depending on all the previous ones.
Decoder The part of a neural network that generates output from an internal representation.
DeepMind A leading AI research lab, now part of Google.
Diffusion Model A generative AI model that creates data by learning to reverse a gradual noising process.