SDXL based text-to-image model applying Distribution Matching Distillation, supporting zero-shot identity generation in 2-5s. https://ai-visionboard.com
