Prompt Caching 與 Batch 投資回報計算器NVIDIA
NVIDIA: Nemotron 3 Ultra Prompt Caching Calculator
Base Input Rate$0.6/M
快取命中讀取價$0.120/M(-80%)
Cache Write SurchargeFree (Automatic)
Output Rate$2.4/M
Simulation Parameters
月度請求量 (Requests)100,000
單次輸入代幣 (Input Tokens)8,000
Prompt 快取命中率 (Cache Hit)75%
單次輸出代幣 (Output Tokens)800
未優化原始成本$672.00
實際淨帳單$384.00
快取節省金額$288.00 (-43%)