Google has launched DiffusionGemma, a new text generation model that operates by converting noise into coherent text, achieving speeds of around 1,000 tokens per second on a single H100 GPU. While it offers impressive speed, the output quality is lower, making it an experimental tool for developers at this stage.