JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators

Researchers have released JuZhou 1.0, a text-to-image foundation model designed specifically for local deployment on edge devices rather than cloud servers. By training the model entirely on Chinese-developed AI hardware, the developers aim to improve user privacy and reduce the latency and costs associated with typical high-end generative AI workloads.