Constructor Detail

Method Detail

• maxCompletionsTokens

   final ScoreModelGrader.SamplingParams.Builder maxCompletionsTokens(Long maxCompletionsTokens)

  The maximum number of tokens the grader model may generate in its response.

• maxCompletionsTokens

   final ScoreModelGrader.SamplingParams.Builder maxCompletionsTokens(Long maxCompletionsTokens)

  Alias for Builder.maxCompletionsTokens.

  This unboxed primitive overload exists for backwards compatibility.

• maxCompletionsTokens

   final ScoreModelGrader.SamplingParams.Builder maxCompletionsTokens(Optional<Long> maxCompletionsTokens)

  Alias for calling Builder.maxCompletionsTokens with maxCompletionsTokens.orElse(null).

• maxCompletionsTokens

   final ScoreModelGrader.SamplingParams.Builder maxCompletionsTokens(JsonField<Long> maxCompletionsTokens)

  Sets Builder.maxCompletionsTokens to an arbitrary JSON value.

  You should usually call Builder.maxCompletionsTokens with a well-typed Long value instead. This method is primarily for setting the field to an undocumented or not yet supported value.

• reasoningEffort

   final ScoreModelGrader.SamplingParams.Builder reasoningEffort(ReasoningEffort reasoningEffort)

  Constrains effort on reasoning for reasoning models. Currently supported values are minimal, low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.

  Note: The gpt-5-pro model defaults to (and only supports) high reasoning effort.

• reasoningEffort

   final ScoreModelGrader.SamplingParams.Builder reasoningEffort(Optional<ReasoningEffort> reasoningEffort)

  Alias for calling Builder.reasoningEffort with reasoningEffort.orElse(null).

• reasoningEffort

   final ScoreModelGrader.SamplingParams.Builder reasoningEffort(JsonField<ReasoningEffort> reasoningEffort)

  Sets Builder.reasoningEffort to an arbitrary JSON value.

  You should usually call Builder.reasoningEffort with a well-typed ReasoningEffort value instead. This method is primarily for setting the field to an undocumented or not yet supported value.

• seed

   final ScoreModelGrader.SamplingParams.Builder seed(Long seed)

  A seed value to initialize the randomness, during sampling.

• seed

   final ScoreModelGrader.SamplingParams.Builder seed(Long seed)

  Alias for Builder.seed.

  This unboxed primitive overload exists for backwards compatibility.

• seed

   final ScoreModelGrader.SamplingParams.Builder seed(Optional<Long> seed)

  Alias for calling Builder.seed with seed.orElse(null).

• seed

   final ScoreModelGrader.SamplingParams.Builder seed(JsonField<Long> seed)

  Sets Builder.seed to an arbitrary JSON value.

  You should usually call Builder.seed with a well-typed Long value instead. This method is primarily for setting the field to an undocumented or not yet supported value.

• temperature

   final ScoreModelGrader.SamplingParams.Builder temperature(Double temperature)

  A higher temperature increases randomness in the outputs.

• temperature

   final ScoreModelGrader.SamplingParams.Builder temperature(Double temperature)

  Alias for Builder.temperature.

  This unboxed primitive overload exists for backwards compatibility.

• temperature

   final ScoreModelGrader.SamplingParams.Builder temperature(Optional<Double> temperature)

  Alias for calling Builder.temperature with temperature.orElse(null).

• temperature

   final ScoreModelGrader.SamplingParams.Builder temperature(JsonField<Double> temperature)

  Sets Builder.temperature to an arbitrary JSON value.

  You should usually call Builder.temperature with a well-typed Double value instead. This method is primarily for setting the field to an undocumented or not yet supported value.

• topP

   final ScoreModelGrader.SamplingParams.Builder topP(Double topP)

  An alternative to temperature for nucleus sampling; 1.0 includes all tokens.

• topP

   final ScoreModelGrader.SamplingParams.Builder topP(Double topP)

  Alias for Builder.topP.

  This unboxed primitive overload exists for backwards compatibility.

• topP

   final ScoreModelGrader.SamplingParams.Builder topP(Optional<Double> topP)

  Alias for calling Builder.topP with topP.orElse(null).

• topP

   final ScoreModelGrader.SamplingParams.Builder topP(JsonField<Double> topP)

  Sets Builder.topP to an arbitrary JSON value.

  You should usually call Builder.topP with a well-typed Double value instead. This method is primarily for setting the field to an undocumented or not yet supported value.

• additionalProperties

   final ScoreModelGrader.SamplingParams.Builder additionalProperties(Map<String, JsonValue> additionalProperties)

• putAdditionalProperty

   final ScoreModelGrader.SamplingParams.Builder putAdditionalProperty(String key, JsonValue value)

• putAllAdditionalProperties

   final ScoreModelGrader.SamplingParams.Builder putAllAdditionalProperties(Map<String, JsonValue> additionalProperties)

• removeAdditionalProperty

   final ScoreModelGrader.SamplingParams.Builder removeAdditionalProperty(String key)

• removeAllAdditionalProperties

   final ScoreModelGrader.SamplingParams.Builder removeAllAdditionalProperties(Set<String> keys)

• build

   final ScoreModelGrader.SamplingParams build()

  Returns an immutable instance of SamplingParams.

  Further updates to this Builder will not mutate the returned instance.