We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
prompt和completion消耗的token都会统计进来
API可以使用的模型版本通常落后于官网