Artificial Analysis said on September 24, 2026, that Ming-Image-0.1-Design led open-weights models in its UI/UX design category, ranking 17th of 81 models there versus 45th of 160 overall. It also ranked the model first among open-weights models for layout and text rendering—19th of 67 and 23rd of 65, respectively, across all models. The evaluator describes it as a 6-billion-parameter text-to-image model from inclusionAI, Ant Group’s AI initiative, built for interfaces, infographics, posters and other text-rich designs. It supports transparent-background output and has open weights under the MIT license. Artificial Analysis tested it at 2K resolution with 12 inference steps, the settings inclusionAI recommends.