Google DeepMind releases DiffusionGemma, a model that runs local AI 4x faster

Google DeepMind released DiffusionGemma, an experimental AI model that processes text locally four times faster than standard Gemma models while maintaining comparable accuracy.