AI TRENDS | OpenAI Engineers Say New System Optimization Cuts AI Inference Costs by More Than Half

OpenAI’s engineering team has told some colleagues that the company has found a new system optimization method that can reduce AI model inference costs by more than half. According to Odaily, inference costs refer to the computing resources consumed when a model runs in practice and responds to user requests. The reported optimization mainly comes from improving the utilization of existing server resources rather than relying on additional investments in new computing chips. The Information reported the development.