中國 AI 模型 GLM-5.3-Flash 以極低成本媲美競爭對手

z.ai released glm-5.3-flash, a 320b parameter model with 18b active parameters, a 1m token context window, and multimodal capabilities. it scores 57 points on the intelligence index, behind glm-5.3's 60 points. glm-5.3-flash runs entirely on chinese ai chips using z.ai's software, achieving efficiency comparable to nvidia gpus. at $0.09 per task, it's 7.5x cheaper than glm-5.3, putting it on the pareto frontier of intelligence and cost. this development adds to growing price pressure from chinese ai models on western providers.