LLMSamplingParams

Sampling and generation parameters for controlling LLM text output

Properties

Name	Type	Description	Notes
maxTokens	Integer	Maximum tokens to generate (>0 if set; provider-dependent limits apply)	[optional]
temperature	Float	Sampling temperature 0.0-2.0 (0.0=deterministic, 2.0=highly random)	[optional]
topP	Float	Nucleus sampling threshold 0.0-1.0 (smaller values focus on higher probability tokens)	[optional]
topK	Integer	Top-k sampling limit (>0 if set; primarily for local/open-source models)	[optional]
frequencyPenalty	Float	Frequency penalty -2.0 to 2.0 (positive values reduce repetition based on frequency)	[optional]
presencePenalty	Float	Presence penalty -2.0 to 2.0 (positive values encourage topic diversity)	[optional]
stopSequences	List<String>	Generation stop sequences (≤10 sequences; each ≤100 chars; generation halts on exact match)	[optional]