kernl

Interface: LanguageModelRequestSettings

Defined in: packages/protocol/src/language-model/request.ts:97

Settings to use when calling an LLM.

This class holds optional model configuration parameters (e.g. temperature, topP, penalties, truncation, etc.).

Not all models/providers support all of these parameters, so please check the API documentation for the specific model and provider you are using.

Properties

PropertyTypeDescriptionDefined in
frequencyPenalty?numberThe frequency penalty to use when calling the model.packages/protocol/src/language-model/request.ts:111
maxTokens?numberThe maximum number of output tokens to generate.packages/protocol/src/language-model/request.ts:137
parallelToolCalls?booleanWhether to use parallel tool calls when calling the model. Defaults to false if not provided.packages/protocol/src/language-model/request.ts:127
presencePenalty?numberThe presence penalty to use when calling the model.packages/protocol/src/language-model/request.ts:116
providerOptions?SharedProviderOptionsAdditional provider specific metadata to be passed directly to the model request.packages/protocol/src/language-model/request.ts:159
reasoning?ModelSettingsReasoningThe reasoning settings to use when calling the model.packages/protocol/src/language-model/request.ts:148
store?booleanWhether to store the generated model response for later retrieval. Defaults to true if not provided.packages/protocol/src/language-model/request.ts:143
temperature?numberThe temperature to use when calling the model.packages/protocol/src/language-model/request.ts:101
text?ModelSettingsTextThe text settings to use when calling the model.packages/protocol/src/language-model/request.ts:153
toolChoice?LanguageModelToolChoiceThe tool choice to use when calling the model.packages/protocol/src/language-model/request.ts:121
topP?numberThe topP to use when calling the model.packages/protocol/src/language-model/request.ts:106
truncation?"auto" | "disabled"The truncation strategy to use when calling the model.packages/protocol/src/language-model/request.ts:132

On this page