Type Alias GenerateTextFn

Type Declaration

• • (
        promptOrOptions:
            | string
            | {
                maxTokens?: number;
                prompt: string;
                system?: { cacheBreakpoint?: boolean; text: string }[];
            },
    ): Promise<string>

  • Parameters

    • promptOrOptions:
          | string
          | {
              maxTokens?: number;
              prompt: string;
              system?: { cacheBreakpoint?: boolean; text: string }[];
          }
      • string
      • {
            maxTokens?: number;
            prompt: string;
            system?: { cacheBreakpoint?: boolean; text: string }[];
        }

        • OptionalmaxTokens?: number

          Upper bound on completion tokens for this call. Caps tail spend when a model misbehaves and yaps beyond the intended output size (provider defaults sit at 4-8k tokens). Use ~2× the typical response size so well-behaved calls finish naturally and only runaway generations hit the cap.

        • prompt: string

        • Optionalsystem?: { cacheBreakpoint?: boolean; text: string }[]

    Returns Promise<string>