OptimizationOptions Class

Package:
com.azure.ai.agents.models
Maven Artifact:
com.azure:azure-ai-agents:2.2.0
  • java.lang.Object
    • com.azure.ai.agents.models.OptimizationOptions

Implements

JsonSerializable<OptimizationOptions>

public final class OptimizationOptions
implements JsonSerializable<OptimizationOptions>

Tuning knobs and run-mode for an optimization job.

Constructor Summary

Constructor Description
OptimizationOptions()

Creates an instance of OptimizationOptions class.

Method Summary

Modifier and Type Method and Description
static OptimizationOptions fromJson(JsonReader jsonReader)

Reads an instance of OptimizationOptions from the JsonReader.
String getEvalModel()

Get the evalModel property: Model deployment used for evaluation.
EvaluationLevel getEvaluationLevel()

Get the evaluationLevel property: Evaluation granularity.
Integer getMaxCandidates()

Get the maxCandidates property: Maximum number of optimization candidates to generate.
Map<String,BinaryData> getOptimizationConfig()

Get the optimizationConfig property: Per-target-attribute configuration overrides.
String getOptimizationModel()

Get the optimizationModel property: Model deployment for optimization reasoning (must be gpt-5 family).
OptimizationOptions setEvalModel(String evalModel)

Set the evalModel property: Model deployment used for evaluation.
OptimizationOptions setEvaluationLevel(EvaluationLevel evaluationLevel)

Set the evaluationLevel property: Evaluation granularity.
OptimizationOptions setMaxCandidates(Integer maxCandidates)

Set the maxCandidates property: Maximum number of optimization candidates to generate.
OptimizationOptions setOptimizationConfig(Map<String,BinaryData> optimizationConfig)

Set the optimizationConfig property: Per-target-attribute configuration overrides.
OptimizationOptions setOptimizationModel(String optimizationModel)

Set the optimizationModel property: Model deployment for optimization reasoning (must be gpt-5 family).
JsonWriter toJson(JsonWriter jsonWriter)

Methods inherited from java.lang.Object

clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Constructor Details

OptimizationOptions

public OptimizationOptions()

Creates an instance of OptimizationOptions class.

Method Details

fromJson

public static OptimizationOptions fromJson(JsonReader jsonReader)

Reads an instance of OptimizationOptions from the JsonReader.

Parameters:

jsonReader - The JsonReader being read.

Returns:

An instance of OptimizationOptions if the JsonReader was pointing to an instance of it, or null if it was pointing to JSON null.

Throws:

IOException

 - If an error occurs while reading the OptimizationOptions.

getEvalModel

public String getEvalModel()

Get the evalModel property: Model deployment used for evaluation. Defaults to server config (typically 'gpt-4o').

Returns:

the evalModel value.

getEvaluationLevel

public EvaluationLevel getEvaluationLevel()

Get the evaluationLevel property: Evaluation granularity. Null/omitted means per-item single-turn. Set to 'conversation' for per-conversation multi-turn simulation scoring.

Returns:

the evaluationLevel value.

getMaxCandidates

public Integer getMaxCandidates()

Get the maxCandidates property: Maximum number of optimization candidates to generate. Must be >= 1. Default: 5.

Returns:

the maxCandidates value.

getOptimizationConfig

public Map<String,BinaryData> getOptimizationConfig()

Get the optimizationConfig property: Per-target-attribute configuration overrides. Contains skills, tools, system_prompt for the agent, plus model space for model optimization.

Returns:

the optimizationConfig value.

getOptimizationModel

public String getOptimizationModel()

Get the optimizationModel property: Model deployment for optimization reasoning (must be gpt-5 family). Falls back to the default eval model when not set.

Returns:

the optimizationModel value.

setEvalModel

public OptimizationOptions setEvalModel(String evalModel)

Set the evalModel property: Model deployment used for evaluation. Defaults to server config (typically 'gpt-4o').

Parameters:

evalModel - the evalModel value to set.

Returns:

the OptimizationOptions object itself.

setEvaluationLevel

public OptimizationOptions setEvaluationLevel(EvaluationLevel evaluationLevel)

Set the evaluationLevel property: Evaluation granularity. Null/omitted means per-item single-turn. Set to 'conversation' for per-conversation multi-turn simulation scoring.

Parameters:

evaluationLevel - the evaluationLevel value to set.

Returns:

the OptimizationOptions object itself.

setMaxCandidates

public OptimizationOptions setMaxCandidates(Integer maxCandidates)

Set the maxCandidates property: Maximum number of optimization candidates to generate. Must be >= 1. Default: 5.

Parameters:

maxCandidates - the maxCandidates value to set.

Returns:

the OptimizationOptions object itself.

setOptimizationConfig

public OptimizationOptions setOptimizationConfig(Map<String,BinaryData> optimizationConfig)

Set the optimizationConfig property: Per-target-attribute configuration overrides. Contains skills, tools, system_prompt for the agent, plus model space for model optimization.

Parameters:

optimizationConfig - the optimizationConfig value to set.

Returns:

the OptimizationOptions object itself.

setOptimizationModel

public OptimizationOptions setOptimizationModel(String optimizationModel)

Set the optimizationModel property: Model deployment for optimization reasoning (must be gpt-5 family). Falls back to the default eval model when not set.

Parameters:

optimizationModel - the optimizationModel value to set.

Returns:

the OptimizationOptions object itself.

toJson

public JsonWriter toJson(JsonWriter jsonWriter)

Parameters:

jsonWriter

Throws:

Applies to