Package: @effect/ai
Module: Response
Represents the reason why a model finished generation of a response.
Possible finish reasons:
"stop": The model generated a stop sequence."length": The model exceeded its token budget."content-filter": The model generated content which violated a content filter."tool-calls": The model triggered a tool call."error": The model encountered an error."pause": The model requested to pause execution."other": The model stopped for a reason not supported by this protocol."unknown": The model did not specify a finish reason.Signature
declare const FinishReason: Schema.Literal<["stop", "length", "content-filter", "tool-calls", "error", "pause", "other", "unknown"]>
Since v1.0.0