SenseTime leads AI infrastructure services, promotes integrated computing and electrical collaboration, and improves inference performance by 4 times

SenseTime, as a leader in domestic AI infrastructure services, has actively promoted the application of innovative technologies such as training-push integration and computing-electricity collaboration, and has achieved remarkable results. Its large-scale devices have achieved a 4-fold increase in the number of requests per second (QPS) under the same computing power and electricity cost in the inference scenario, and have achieved elastic on-demand scaling of inference services, optimizing the overall cost of large-scale AI inference. The total computing power scale of SenseTime's large-scale devices has exceeded 20,000 PetaFLOPS, providing the industry with green intelligent computing services with high energy efficiency.