OUI-1: world's first model for Generative UI

OUI-1 is a finetuned DiffusionGemma model that generates user interfaces locally on consumer GPUs, scoring 71.7% on the Generative UI Benchmark — up from DiffusionGemma's 13.0%. Using a two-phase approach of supervised fine-tuning and self-distillation with a parser-based reward signal, the team resolved a speed/accuracy tradeoff and outperformed larger dense models like Gemma 4 31B, all with just 4B active parameters.