JD Cloud plans to build a computing cluster powered by 100,000 general-purpose GPUs from Chinese chipmaker Moore Threads. Announced at JD’s 2026 Global Technology Explorers Conference, the project will support large-model training and inference as well as embodied AI, with computing capacity available to companies across industries.
JD Cloud said the project would create large-scale computing infrastructure based on Chinese hardware and help bring physical AI into commercial use. The company described it as the first deployment of Chinese GPUs in a 100,000-GPU core computing cluster at a major domestic AI cloud provider. [Securities Times, in Chinese]