Qwen/Qwen3.8-Flash lists $0.16 input, $0.47 output, and $0.016 cache read per million tokens, and z-ai/glm-5.3-flash lists $0.15, $0.50, and $0.03; the official pricing page documents no cache-write rate for either model.