--- title: "GET /v1/projects/{projectsId}/locations/{locationsId}/evaluationRuns/{evaluationRunsId}" method: GET path: "/v1/projects/{projectsId}/locations/{locationsId}/evaluationRuns/{evaluationRunsId}" tags: ["projects"] --- # GET /v1/projects/{projectsId}/locations/{locationsId}/evaluationRuns/{evaluationRunsId} `GET /v1/projects/{projectsId}/locations/{locationsId}/evaluationRuns/{evaluationRunsId}` ## Request - Base URL: `https://aiplatform.googleapis.com` - URL: `https://aiplatform.googleapis.com/v1/projects/{projectsId}/locations/{locationsId}/evaluationRuns/{evaluationRunsId}` - Auth: OAuth 2 and OAuth 2 - Implicit: `https://accounts.google.com/o/oauth2/auth` - Scopes: `https://www.googleapis.com/auth/cloud-platform` - Authorization code: `https://accounts.google.com/o/oauth2/auth` - Token URL: `https://accounts.google.com/o/oauth2/token` scopes: `https://www.googleapis.com/auth/cloud-platform` Gets an Evaluation Run. ## Path parameters - `projectsId` string, required - `locationsId` string, required - `evaluationRunsId` string, required ## Response `200` Successful response - GoogleCloudAiplatformV1EvaluationRun: EvaluationRun is a resource that represents a single evaluation run, which includes a set of prompts, model responses, evaluation configuration and the resulting metrics. - `completionTime` string, google-datetime: Output only. Time when the evaluation run was completed. - `createTime` string, google-datetime: Output only. Time when the evaluation run was created. - `dataSource` GoogleCloudAiplatformV1EvaluationRunDataSource: The data source for the evaluation run. - `bigqueryRequestSet` GoogleCloudAiplatformV1BigQueryRequestSet: The request set for the evaluation run. - `candidateResponseColumns` object: Optional. Map of candidate name to candidate response column name. The column will be in evaluation_item.CandidateResponse format. - `promptColumn` string: Optional. The name of the column that contains the requests to evaluate. This will be in evaluation_item.EvalPrompt format. - `rubricsColumn` string: Optional. The name of the column that contains the rubrics. This is in evaluation_rubric.RubricGroup format. - `samplingConfig` GoogleCloudAiplatformV1BigQueryRequestSetSamplingConfig: The sampling config. - `samplingCount` integer: Optional. The total number of logged data to import. If available data is less than the sampling count, all data will be imported. Default is 100. - `samplingDuration` string, google-duration: Optional. How long to wait before sampling data from the BigQuery table. If not specified, defaults to 0. - `samplingMethod` 'SAMPLING_METHOD_UNSPECIFIED' | 'RANDOM': Optional. The sampling method to use. - `uri` string: Required. The URI of a BigQuery table. e.g. bq://projectId.bqDatasetId.bqTableId - `evaluationSet` string: The EvaluationSet resource name. Format: `projects/{project}/locations/{location}/evaluationSets/{evaluation_set}` - `displayName` string: Required. The display name of the Evaluation Run. - `encryptionSpec` GoogleCloudAiplatformV1EncryptionSpec: Represents a customer-managed encryption key specification that can be applied to a Vertex AI resource. - `kmsKeyName` string: Required. Resource name of the Cloud KMS key used to protect the resource. The Cloud KMS key must be in the same region as the resource. It must have the format `projects/{project}/locations/{location}/keyRings/{key_ring}/cryptoKeys/{crypto_key}`. - `error` GoogleRpcStatus: The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). - `code` integer: The status code, which should be an enum value of google.rpc.Code. - `details` object[]: A list of messages that carry the error details. There is a common set of message types for APIs to use. - `message` string: A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client. - `evaluationConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfig: The Evalution configuration used for the evaluation run. - `allowCrossRegionModel` boolean: Optional. Allows the evaluation run to use cross region models. When this flag is set, the service may route traffic to other regions if a model is unavailable in the current region (e.g., to a `global` endpoint). If a fully-qualified model endpoint resource name with a different region than the run location is provided elsewhere in the run config, this flag must be set to true or the request will fail. - `autoraterConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - `audioTimestamp` boolean: Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - `audioTranscriptionConfig` GoogleCloudAiplatformV1AudioTranscriptionConfig: Configuration for speech recognition (transcription). - `adaptationPhrases` string[]: Optional. Deprecated: Use `custom_vocabulary` instead. A list of phrases to bias the speech recognition model towards. - `customVocabulary` string[]: Optional. A list of custom vocabulary phrases to bias the speech recognition model toward recognizing specific terms. - `diarization` boolean: Optional. Configures speaker diarization. - `languageAuto` GoogleCloudAiplatformV1AudioTranscriptionConfigLanguageAuto: Deprecated: Use top-level `language_codes` instead. Indicates the language of the audio should be automatically detected. - `languageCodes` string[]: Optional. BCP-47 language codes providing hints about the languages present in the audio. If omitted or empty, defaults to automatic language detection. - `languageHints` GoogleCloudAiplatformV1AudioTranscriptionConfigLanguageHints: Deprecated: Use top-level `language_codes` instead. Provides hints to the model about possible languages present in the audio. - `languageCodes` string[]: Required. Deprecated: Use top-level `language_codes` instead. BCP-47 language codes. At least one must be specified. - `mode` 'MODE_UNSPECIFIED' | 'VERBATIM' | 'SMART': Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`. - `sourceText` string: Optional. The source text to be edited or rewritten via voice instructions. When provided in `SMART` mode, the model evaluates the audio input against this text and emits the complete replacement text in the candidate response rather than appending to it. - `wordTimestamp` boolean: Optional. Configures word-level timestamp generation. - `candidateCount` integer: Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - `enableAffectiveDialog` boolean: Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - `frequencyPenalty` number, float: Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - `imageConfig` GoogleCloudAiplatformV1ImageConfig: Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. - `aspectRatio` string: Optional. The desired aspect ratio for the generated images. The following aspect ratios are supported: "1:1" "2:3", "3:2" "3:4", "4:3" "4:5", "5:4" "9:16", "16:9" "21:9" - `imageOutputOptions` GoogleCloudAiplatformV1ImageConfigImageOutputOptions: The image output format for generated images. - `compressionQuality` integer: Optional. The compression quality of the output image. - `mimeType` string: Optional. The image format that the output should be saved as. - `imageSize` string: Optional. Specifies the size of generated images. Supported values are `1K`, `2K`, `4K`. If not specified, the model will use default value `1K`. - `personGeneration` 'PERSON_GENERATION_UNSPECIFIED' | 'ALLOW_ALL' | 'ALLOW_ADULT' | 'ALLOW_NONE': Optional. Controls whether the model can generate people. - `prominentPeople` 'PROMINENT_PEOPLE_UNSPECIFIED' | 'ALLOW_PROMINENT_PEOPLE' | 'BLOCK_PROMINENT_PEOPLE': Optional. Controls whether prominent people (celebrities) generation is allowed. If used with personGeneration, personGeneration enum would take precedence. For instance, if ALLOW_NONE is set, all person generation would be blocked. If this field is unspecified, the default behavior is to allow prominent people. - `logprobs` integer: Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - `maxOutputTokens` integer: Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - `mediaResolution` 'MEDIA_RESOLUTION_UNSPECIFIED' | 'MEDIA_RESOLUTION_LOW' | 'MEDIA_RESOLUTION_MEDIUM' | 'MEDIA_RESOLUTION_HIGH': Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - `presencePenalty` number, float: Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - `responseFormat` GoogleCloudAiplatformV1ResponseFormat[]: Optional. New response format field for the model to configure output formatting and delivery. - `audio` GoogleCloudAiplatformV1AudioResponseFormat: Configuration for audio-specific output formatting. - `bitRate` integer: Optional. Bit rate in bits per second (bps). Only applicable for compressed formats (MP3, Opus). - `delivery` 'DELIVERY_UNSPECIFIED' | 'INLINE' | 'URI': Optional. Delivery mode for the generated content. - `mimeType` 'MIME_TYPE_UNSPECIFIED' | 'AUDIO_MP3' | 'AUDIO_OGG_OPUS' | 'AUDIO_L16' | 'AUDIO_WAV' | 'AUDIO_ALAW' | 'AUDIO_MULAW': Optional. The MIME type of the audio output. - `sampleRate` integer: Optional. Sample rate for the generated audio in Hertz. - `image` GoogleCloudAiplatformV1ImageResponseFormat: Configuration for image-specific output formatting. - `aspectRatio` 'ASPECT_RATIO_UNSPECIFIED' | 'ASPECT_RATIO_ONE_BY_ONE' | 'ASPECT_RATIO_TWO_BY_THREE' | 'ASPECT_RATIO_THREE_BY_TWO' | 'ASPECT_RATIO_THREE_BY_FOUR' | 'ASPECT_RATIO_FOUR_BY_THREE' | 'ASPECT_RATIO_FOUR_BY_FIVE' | 'ASPECT_RATIO_FIVE_BY_FOUR' | 'ASPECT_RATIO_NINE_BY_SIXTEEN' | 'ASPECT_RATIO_SIXTEEN_BY_NINE' | 'ASPECT_RATIO_TWENTY_ONE_BY_NINE' | 'ASPECT_RATIO_ONE_BY_EIGHT' | 'ASPECT_RATIO_EIGHT_BY_ONE' | 'ASPECT_RATIO_ONE_BY_FOUR' | 'ASPECT_RATIO_FOUR_BY_ONE': Optional. The aspect ratio for the image output. - `delivery` 'DELIVERY_UNSPECIFIED' | 'INLINE' | 'URI': Optional. Delivery mode for the generated content. - `imageSize` 'IMAGE_SIZE_UNSPECIFIED' | 'IMAGE_SIZE_FIVE_TWELVE' | 'IMAGE_SIZE_ONE_K' | 'IMAGE_SIZE_TWO_K' | 'IMAGE_SIZE_FOUR_K': Optional. The size of the image output. - `mimeType` 'MIME_TYPE_UNSPECIFIED' | 'IMAGE_JPEG': Optional. The MIME type of the image output. - `text` GoogleCloudAiplatformV1TextResponseFormat: Configuration for text-specific output formatting. - `mimeType` 'MIME_TYPE_UNSPECIFIED' | 'APPLICATION_JSON' | 'TEXT_PLAIN': Optional. The IANA standard MIME type of the response. - `schema` unknown - `video` GoogleCloudAiplatformV1VideoResponseFormat: Configuration for video-specific output formatting. - `aspectRatio` 'ASPECT_RATIO_UNSPECIFIED' | 'ASPECT_RATIO_SIXTEEN_BY_NINE' | 'ASPECT_RATIO_NINE_BY_SIXTEEN': The aspect ratio for the video output. - `delivery` 'DELIVERY_UNSPECIFIED' | 'INLINE' | 'URI': Optional. Delivery mode for the generated content. - `duration` string, google-duration: Optional. The duration for the video output. - `gcsUri` string: Optional. The Google Cloud Storage URI to store the video output. Required for Vertex if delivery is URI. - `resolution` string: Optional. The video output resolution. Supported values: "360p", "720p", "1080p", "4k". - `responseJsonSchema` unknown - `responseLogprobs` boolean: Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - `responseMimeType` string: Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - `responseModalities` string[]: Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - `responseSchema` GoogleCloudAiplatformV1Schema: Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). - `additionalProperties` unknown - `anyOf` GoogleCloudAiplatformV1Schema[]: Optional. The instance must be valid against any (one or more) of the subschemas listed in `any_of`. - `default` unknown - `defs` object: Optional. `defs` provides a map of schema definitions that can be reused by `ref` elsewhere in the schema. Only allowed at root level of the schema. - `description` string: Optional. Describes the data. The model uses this field to understand the purpose of the schema and how to use it. It is a best practice to provide a clear and descriptive explanation for the schema and its properties here, rather than in the prompt. - `enum` string[]: Optional. Possible values of the field. This field can be used to restrict a value to a fixed set of values. To mark a field as an enum, set `format` to `enum` and provide the list of possible values in `enum`. For example: 1. To define directions: `{type:STRING, format:enum, enum:["EAST", "NORTH", "SOUTH", "WEST"]}` 2. To define apartment numbers: `{type:INTEGER, format:enum, enum:["101", "201", "301"]}` - `example` unknown - `format` string: Optional. The format of the data. For `NUMBER` type, format can be `float` or `double`. For `INTEGER` type, format can be `int32` or `int64`. For `STRING` type, format can be `email`, `byte`, `date`, `date-time`, `password`, and other formats to further refine the data type. - `items` GoogleCloudAiplatformV1Schema (recursive) - `maxItems` string, int64: Optional. If type is `ARRAY`, `max_items` specifies the maximum number of items in an array. - `maxLength` string, int64: Optional. If type is `STRING`, `max_length` specifies the maximum length of the string. - `maxProperties` string, int64: Optional. If type is `OBJECT`, `max_properties` specifies the maximum number of properties that can be provided. - `maximum` number, double: Optional. If type is `INTEGER` or `NUMBER`, `maximum` specifies the maximum allowed value. - `minItems` string, int64: Optional. If type is `ARRAY`, `min_items` specifies the minimum number of items in an array. - `minLength` string, int64: Optional. If type is `STRING`, `min_length` specifies the minimum length of the string. - `minProperties` string, int64: Optional. If type is `OBJECT`, `min_properties` specifies the minimum number of properties that can be provided. - `minimum` number, double: Optional. If type is `INTEGER` or `NUMBER`, `minimum` specifies the minimum allowed value. - `nullable` boolean: Optional. Indicates if the value of this field can be null. - `pattern` string: Optional. If type is `STRING`, `pattern` specifies a regular expression that the string must match. - `properties` object: Optional. If type is `OBJECT`, `properties` is a map of property names to schema definitions for each property of the object. - `propertyOrdering` string[]: Optional. Order of properties displayed or used where order matters. This is not a standard field in OpenAPI specification, but can be used to control the order of properties. - `ref` string: Optional. Allows referencing another schema definition to use in place of this schema. The value must be a valid reference to a schema in `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring - `required` string[]: Optional. If type is `OBJECT`, `required` lists the names of properties that must be present. - `title` string: Optional. Title for the schema. - `type` 'TYPE_UNSPECIFIED' | 'STRING' | 'NUMBER' | 'INTEGER' | 'BOOLEAN' | 'ARRAY' | 'OBJECT' | 'NULL': Optional. Data type of the schema field. - `routingConfig` GoogleCloudAiplatformV1GenerationConfigRoutingConfig: The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. - `autoMode` GoogleCloudAiplatformV1GenerationConfigRoutingConfigAutoRoutingMode: The configuration for automated routing. When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference. - `modelRoutingPreference` 'UNKNOWN' | 'PRIORITIZE_QUALITY' | 'BALANCED' | 'PRIORITIZE_COST': The model routing preference. - `manualMode` GoogleCloudAiplatformV1GenerationConfigRoutingConfigManualRoutingMode: The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. - `modelName` string: The name of the model to use. Only public LLM models are accepted. - `seed` integer: Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - `speechConfig` GoogleCloudAiplatformV1SpeechConfig: Configuration for speech generation. - `languageCode` string: Optional. The language code (ISO 639-1) for the speech synthesis. - `multiSpeakerVoiceConfig` GoogleCloudAiplatformV1MultiSpeakerVoiceConfig: Configuration for a multi-speaker text-to-speech request. - `speakerVoiceConfigs` GoogleCloudAiplatformV1SpeakerVoiceConfig[]: Required. A list of configurations for the voices of the speakers. Exactly two speaker voice configurations must be provided. - … - `voiceConfig` GoogleCloudAiplatformV1VoiceConfig: Configuration for a voice. - `prebuiltVoiceConfig` GoogleCloudAiplatformV1PrebuiltVoiceConfig: Configuration for a prebuilt voice. - … - `replicatedVoiceConfig` GoogleCloudAiplatformV1ReplicatedVoiceConfig: The configuration for the replicated voice to use. - … - `voice` string: Optional. The speaker identifier for synthesis. Supported formats: * Speaker name for prebuilt voices (for example, `Orus` or `Kore`). * Voice ID for stored voices (for example, `voice_xxx`). * Voice replication key (for example, `voicekey_xxx`). - `stopSequences` string[]: Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - `temperature` number, float: Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - `thinkingConfig` GoogleCloudAiplatformV1GenerationConfigThinkingConfig: Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. - `includeThoughts` boolean: Optional. If true, the model will include its thoughts in the response. "Thoughts" are the intermediate steps the model takes to arrive at the final response. They can provide insights into the model's reasoning process and help with debugging. If this is true, thoughts are returned only when available. - `thinkingBudget` integer: Optional. The token budget for the model's thinking process. The model will make a best effort to stay within this budget. This can be used to control the trade-off between response quality and latency. - `thinkingLevel` 'THINKING_LEVEL_UNSPECIFIED' | 'LOW' | 'MEDIUM' | 'HIGH' | 'MINIMAL': Optional. The number of thoughts tokens that the model should generate. - `topK` number, float: Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - `topP` number, float: Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - `translationConfig` GoogleCloudAiplatformV1TranslationConfig: Config for translation features. - `echoTargetLanguage` boolean: Optional. If `true`, the model will generate audio when the target language is spoken, essentially it will parrot the input. If `false`, we will not produce audio for the target language. - `targetLanguageCode` string: Required. The target language for translation. Supported values are BCP-47 language codes (e.g. "en", "es", "fr"). - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `cloudLoggingConfig` GoogleCloudAiplatformV1CloudLoggingConfig: Specifies configuration for exporting evaluation results to Cloud Logging. - `project` string: Optional. Google Cloud project to write logs to. Defaults to the request project. - `resourceLabels` object: Optional. MonitoredResource labels to associate the log with. The backend will automatically inject project and location. - `resourceType` string: Optional. MonitoredResource type. Defaults to "global" if unspecified. - `tracingContext` GoogleCloudAiplatformV1CloudLoggingConfigTracingContext: Tracing context for Observability correlation. - `conversationId` string: Optional. Unique identifier for a conversation (session thread), used to store and correlate messages within a conversation. The value corresponds to the `gen_ai.conversation.id` field in the the OpenTelemetry GenAI attributes. - `spanId` string: Optional. ID of the Cloud Trace span associated with the current operation in which the log is being written. e.g., `7a2190356c3fc94b`. If a span is being evaluated, this field should be populated. - `traceId` string: Optional. Trace ID being written to Cloud Trace in association with this log entry. e.g., `12345`, the numeric ID from the resource name. If a trace or span is being evaluated, this field should be populated. - `datasetCustomMetrics` GoogleCloudAiplatformV1DatasetCustomMetric[]: Optional. Specifications for custom dataset-level aggregations. - `aggregationFunction` string: Required. The Python code string containing the aggregation function. Expected function signature: `def aggregate(instances: list[dict[str, Any]]) -> dict[str, float]:` The `instances` argument is a list of dictionaries, where each dictionary represents a single evaluation result item. The structure of each dictionary corresponds to the fields in the `EvaluationResult` message. This includes: - `"request"`: Contains the original input data and model inputs (from `EvaluationResult.EvaluationRequest`). - `"candidate_results"`: Contains the results of any instance-level metrics (from `EvaluationResult.CandidateResults`). Example of a single item in the `instances` list: { "request": { "prompt": {"text": "What is the capital of France?"}, "golden_response": {"text": "Paris"}, "candidate_responses": [{"candidate": "model-v1", "text": "Paris"}] }, "candidate_results": [ {"metric": "exact_match", "score": 1.0}, {"metric": "bleu", "score": 0.9} ] } - `displayName` string: Optional. A display name for this custom summary metric. Used to prefix keys in the output summaryMetrics map. If not provided, a default name like "dataset_custom_metric_1", "dataset_custom_metric_2", etc., will be generated based on the order in the repeated field. - `lossAnalysisConfig` GoogleCloudAiplatformV1LossAnalysisConfig[]: Optional. Specifications for loss analysis. Each config can be specified for one metric. - `candidate` string: Required. The candidate model/agent to analyze (e.g., "gemini-3.0-pro"). This targets the specific CandidateResult within the EvaluationResult. - `metric` string: Required. The metric to analyze (e.g., "tool_use_quality"). This filters the EvaluationItems in the EvalSet to only those where EvaluationResult.metric matches this value. - `metrics` GoogleCloudAiplatformV1EvaluationRunMetric[]: Optional. The metrics to be calculated in the evaluation run. Required when analysis_configs is not set. - `computationBasedMetricSpec` GoogleCloudAiplatformV1EvaluationRunMetricComputationBasedMetricSpec: Specification for a computation based metric. - `parameters` object: Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - `type` 'COMPUTATION_BASED_METRIC_TYPE_UNSPECIFIED' | 'EXACT_MATCH' | 'BLEU' | 'ROUGE': Required. The type of the computation based metric. - `llmBasedMetricSpec` GoogleCloudAiplatformV1EvaluationRunMetricLLMBasedMetricSpec: Specification for an LLM based metric. - `additionalConfig` object: Optional. Optional additional configuration for the metric. - `judgeAutoraterConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - `audioTimestamp` boolean: Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - `audioTranscriptionConfig` GoogleCloudAiplatformV1AudioTranscriptionConfig: Configuration for speech recognition (transcription). - … - `candidateCount` integer: Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - `enableAffectiveDialog` boolean: Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - `frequencyPenalty` number, float: Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - `imageConfig` GoogleCloudAiplatformV1ImageConfig: Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. - … - `logprobs` integer: Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - `maxOutputTokens` integer: Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - `mediaResolution` 'MEDIA_RESOLUTION_UNSPECIFIED' | 'MEDIA_RESOLUTION_LOW' | 'MEDIA_RESOLUTION_MEDIUM' | 'MEDIA_RESOLUTION_HIGH': Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - `presencePenalty` number, float: Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - `responseFormat` GoogleCloudAiplatformV1ResponseFormat[]: Optional. New response format field for the model to configure output formatting and delivery. - … - `responseJsonSchema` unknown - `responseLogprobs` boolean: Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - `responseMimeType` string: Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - `responseModalities` string[]: Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - `responseSchema` GoogleCloudAiplatformV1Schema: Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). - … - `routingConfig` GoogleCloudAiplatformV1GenerationConfigRoutingConfig: The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. - … - `seed` integer: Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - `speechConfig` GoogleCloudAiplatformV1SpeechConfig: Configuration for speech generation. - … - `stopSequences` string[]: Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - `temperature` number, float: Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - `thinkingConfig` GoogleCloudAiplatformV1GenerationConfigThinkingConfig: Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. - … - `topK` number, float: Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - `topP` number, float: Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - `translationConfig` GoogleCloudAiplatformV1TranslationConfig: Config for translation features. - … - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `metricPromptTemplate` string: Required. Template for the prompt sent to the judge model. - `predefinedRubricGenerationSpec` GoogleCloudAiplatformV1EvaluationRunMetricPredefinedMetricSpec: Specification for a pre-defined metric. - `metricSpecName` string: Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - `parameters` object: Optional. The parameters needed to run the pre-defined metric. - `rubricGenerationSpec` GoogleCloudAiplatformV1EvaluationRunMetricRubricGenerationSpec: Specification for how rubrics should be generated. - `metricResourceName` string: Optional. Resource name of the metric definition. - `modelConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - … - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `promptTemplate` string: Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - `rubricContentType` 'RUBRIC_CONTENT_TYPE_UNSPECIFIED' | 'PROPERTY' | 'NL_QUESTION_ANSWER' | 'PYTHON_CODE_ASSERTION': Optional. The type of rubric content to be generated. - `rubricTypeOntology` string[]: Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - `rubricGroupKey` string: Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - `systemInstruction` string: Optional. System instructions for the judge model. - `metric` string: Required. The name of the metric. - `metricConfig` GoogleCloudAiplatformV1Metric: The metric used for running evaluations. - `aggregationMetrics` string[]: Optional. The aggregation metrics to use. - `bleuSpec` GoogleCloudAiplatformV1BleuSpec: Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1. - `useEffectiveOrder` boolean: Optional. Whether to use_effective_order to compute bleu score. - `computationBasedMetricSpec` GoogleCloudAiplatformV1ComputationBasedMetricSpec: Specification for a computation based metric. - `parameters` object: Optional. A map of parameters for the metric, e.g. {"rouge_type": "rougeL"}. - `type` 'COMPUTATION_BASED_METRIC_TYPE_UNSPECIFIED' | 'EXACT_MATCH' | 'BLEU' | 'ROUGE': Required. The type of the computation based metric. - `customCodeExecutionSpec` GoogleCloudAiplatformV1CustomCodeExecutionSpec: Specificies a metric that is populated by evaluating user-defined Python code. - `evaluationFunction` string: Required. Python function. Expected user to define the following function, e.g.: def evaluate(instance: dict[str, Any]) -> float: Please include this function signature in the code snippet. Instance is the evaluation instance, any fields populated in the instance are available to the function as instance[field_name]. Example: Example input: ``` instance= EvaluationInstance( response=EvaluationInstance.InstanceData(text="The answer is 4."), reference=EvaluationInstance.InstanceData(text="4") ) ``` Example converted input: ``` { 'response': {'text': 'The answer is 4.'}, 'reference': {'text': '4'} } ``` Example python function: ``` def evaluate(instance: dict[str, Any]) -> float: if instance'response' == instance'reference': return 1.0 return 0.0 ``` CustomCodeExecutionSpec is also supported in Batch Evaluation (EvalDataset RPC) and Tuning Evaluation. Each line in the input jsonl file will be converted to dict[str, Any] and passed to the evaluation function. - `exactMatchSpec` GoogleCloudAiplatformV1ExactMatchSpec: Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. - `llmBasedMetricSpec` GoogleCloudAiplatformV1LLMBasedMetricSpec: Specification for an LLM based metric. - `additionalConfig` object: Optional. Optional additional configuration for the metric. - `judgeAutoraterConfig` GoogleCloudAiplatformV1AutoraterConfig: The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `flipEnabled` boolean: Optional. Default is true. Whether to flip the candidate and baseline responses. This is only applicable to the pairwise metric. If enabled, also provide PairwiseMetricSpec.candidate_response_field_name and PairwiseMetricSpec.baseline_response_field_name. When rendering PairwiseMetricSpec.metric_prompt_template, the candidate and baseline fields will be flipped for half of the samples to reduce bias. - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - … - `samplingCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `metricPromptTemplate` string: Required. Template for the prompt sent to the judge model. - `predefinedRubricGenerationSpec` GoogleCloudAiplatformV1PredefinedMetricSpec: The spec for a pre-defined metric. - `metricSpecName` string: Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - `metricSpecParameters` object: Optional. The parameters needed to run the pre-defined metric. - `resultParserConfig` GoogleCloudAiplatformV1EvaluationParserConfig: Config for parsing LLM responses. It can be used to parse the LLM response to be evaluated, or the LLM response from LLM-based metrics/Autoraters. - `customCodeParserConfig` GoogleCloudAiplatformV1EvaluationParserConfigCustomCodeParserConfig: Configuration for parsing the LLM response using custom code. - … - `rubricGenerationSpec` GoogleCloudAiplatformV1RubricGenerationSpec: Specification for how rubrics should be generated. - `modelConfig` GoogleCloudAiplatformV1AutoraterConfig: The configs for autorater. This is applicable to both EvaluateInstances and EvaluateDataset. - … - `promptTemplate` string: Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - `rubricContentType` 'RUBRIC_CONTENT_TYPE_UNSPECIFIED' | 'PROPERTY' | 'NL_QUESTION_ANSWER' | 'PYTHON_CODE_ASSERTION': The type of rubric content to be generated. - `rubricTypeOntology` string[]: Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - `rubricGroupKey` string: Use a pre-defined group of rubrics associated with the input. Refers to a key in the rubric_groups map of EvaluationInstance. - `systemInstruction` string: Optional. System instructions for the judge model. - `metadata` GoogleCloudAiplatformV1MetricMetadata: Metadata about the metric, used for visualization and organization. - `otherMetadata` object: Optional. Flexible metadata for user-defined attributes. - `scoreRange` GoogleCloudAiplatformV1MetricMetadataScoreRange: The range of possible scores for this metric, used for plotting. - `description` string: Optional. The description of the score explaining the directionality etc. - `max` number, double: Required. The maximum value of the score range (inclusive). - `min` number, double: Required. The minimum value of the score range (inclusive). - `step` number, double: Optional. The distance between discrete steps in the range. If unset, the range is assumed to be continuous. - `title` string: Optional. The user-friendly name for the metric. If not set for a registered metric, it will default to the metric's display name. - `pairwiseMetricSpec` GoogleCloudAiplatformV1PairwiseMetricSpec: Spec for pairwise metric. - `baselineResponseFieldName` string: Optional. The field name of the baseline response. - `candidateResponseFieldName` string: Optional. The field name of the candidate response. - `customOutputFormatConfig` GoogleCloudAiplatformV1CustomOutputFormatConfig: Spec for custom output format configuration. - `returnRawOutput` boolean: Optional. Whether to return raw output. - `metricPromptTemplate` string: Required. Metric prompt template for pairwise metric. - `systemInstruction` string: Optional. System instructions for pairwise metric. - `pointwiseMetricSpec` GoogleCloudAiplatformV1PointwiseMetricSpec: Spec for pointwise metric. - `customOutputFormatConfig` GoogleCloudAiplatformV1CustomOutputFormatConfig: Spec for custom output format configuration. - `returnRawOutput` boolean: Optional. Whether to return raw output. - `metricPromptTemplate` string: Required. Metric prompt template for pointwise metric. - `systemInstruction` string: Optional. System instructions for pointwise metric. - `predefinedMetricSpec` GoogleCloudAiplatformV1PredefinedMetricSpec: The spec for a pre-defined metric. - `metricSpecName` string: Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - `metricSpecParameters` object: Optional. The parameters needed to run the pre-defined metric. - `rougeSpec` GoogleCloudAiplatformV1RougeSpec: Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1. - `rougeType` string: Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum. - `splitSummaries` boolean: Optional. Whether to split summaries while using rougeLsum. - `useStemmer` boolean: Optional. Whether to use stemmer to compute rouge score. - `metricResourceName` string: Optional. The resource name of the metric definition. - `predefinedMetricSpec` GoogleCloudAiplatformV1EvaluationRunMetricPredefinedMetricSpec: Specification for a pre-defined metric. - `metricSpecName` string: Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - `parameters` object: Optional. The parameters needed to run the pre-defined metric. - `rubricBasedMetricSpec` GoogleCloudAiplatformV1EvaluationRunMetricRubricBasedMetricSpec: Specification for a metric that is based on rubrics. - `inlineRubrics` GoogleCloudAiplatformV1EvaluationRunMetricRubricBasedMetricSpecRepeatedRubrics: Defines a list of rubrics, used when providing rubrics inline. - `rubrics` GoogleCloudAiplatformV1Rubric[]: The list of rubrics. - `content` GoogleCloudAiplatformV1RubricContent: Content of the rubric, defining the testable criteria. - … - `importance` 'IMPORTANCE_UNSPECIFIED' | 'HIGH' | 'MEDIUM' | 'LOW': Optional. The relative importance of this rubric. - `rubricId` string: Unique identifier for the rubric. This ID is used to refer to this rubric, e.g., in RubricVerdict. - `type` string: Optional. A type designator for the rubric, which can inform how it's evaluated or interpreted by systems or users. It's recommended to use consistent, well-defined, upper snake_case strings. Examples: "SUMMARIZATION_QUALITY", "SAFETY_HARMFUL_CONTENT", "INSTRUCTION_ADHERENCE". - `judgeAutoraterConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - `audioTimestamp` boolean: Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - `audioTranscriptionConfig` GoogleCloudAiplatformV1AudioTranscriptionConfig: Configuration for speech recognition (transcription). - … - `candidateCount` integer: Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - `enableAffectiveDialog` boolean: Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - `frequencyPenalty` number, float: Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - `imageConfig` GoogleCloudAiplatformV1ImageConfig: Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. - … - `logprobs` integer: Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - `maxOutputTokens` integer: Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - `mediaResolution` 'MEDIA_RESOLUTION_UNSPECIFIED' | 'MEDIA_RESOLUTION_LOW' | 'MEDIA_RESOLUTION_MEDIUM' | 'MEDIA_RESOLUTION_HIGH': Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - `presencePenalty` number, float: Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - `responseFormat` GoogleCloudAiplatformV1ResponseFormat[]: Optional. New response format field for the model to configure output formatting and delivery. - … - `responseJsonSchema` unknown - `responseLogprobs` boolean: Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - `responseMimeType` string: Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - `responseModalities` string[]: Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - `responseSchema` GoogleCloudAiplatformV1Schema: Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). - … - `routingConfig` GoogleCloudAiplatformV1GenerationConfigRoutingConfig: The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. - … - `seed` integer: Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - `speechConfig` GoogleCloudAiplatformV1SpeechConfig: Configuration for speech generation. - … - `stopSequences` string[]: Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - `temperature` number, float: Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - `thinkingConfig` GoogleCloudAiplatformV1GenerationConfigThinkingConfig: Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. - … - `topK` number, float: Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - `topP` number, float: Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - `translationConfig` GoogleCloudAiplatformV1TranslationConfig: Config for translation features. - … - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `metricPromptTemplate` string: Optional. Template for the prompt used by the judge model to evaluate against rubrics. - `rubricGenerationSpec` GoogleCloudAiplatformV1EvaluationRunMetricRubricGenerationSpec: Specification for how rubrics should be generated. - `metricResourceName` string: Optional. Resource name of the metric definition. - `modelConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - … - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `promptTemplate` string: Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - `rubricContentType` 'RUBRIC_CONTENT_TYPE_UNSPECIFIED' | 'PROPERTY' | 'NL_QUESTION_ANSWER' | 'PYTHON_CODE_ASSERTION': Optional. The type of rubric content to be generated. - `rubricTypeOntology` string[]: Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - `rubricGroupKey` string: Use a pre-defined group of rubrics associated with the input content. This refers to a key in the `rubric_groups` map of `RubricEnhancedContents`. - `outputConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigOutputConfig: The output config for the evaluation run. - `bigqueryDestination` GoogleCloudAiplatformV1BigQueryDestination: The BigQuery location for the output content. - `outputUri` string: Required. BigQuery URI to a project or table, up to 2000 characters long. When only the project is specified, the Dataset and Table is created. When the full table reference is specified, the Dataset must exist and table must not exist. Accepted forms: * BigQuery path. For example: `bq://projectId` or `bq://projectId.bqDatasetId` or `bq://projectId.bqDatasetId.bqTableId`. - `gcsDestination` GoogleCloudAiplatformV1GcsDestination: The Google Cloud Storage location where the output is to be written to. - `outputUriPrefix` string: Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist. - `promptTemplate` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigPromptTemplate: Prompt template used for inference. - `gcsUri` string: Prompt template stored in Cloud Storage. Format: "gs://my-bucket/file-name.txt". - `promptTemplate` string: Inline prompt template. Template variables should be in the format "{var_name}". Example: "Translate the following from {source_lang} to {target_lang}: {text}" - `rubricConfigs` GoogleCloudAiplatformV1EvaluationRubricConfig[]: Optional. The rubric configs for the evaluation run. They are used to generate rubrics which can be used by rubric-based metrics. Multiple rubric configs can be specified for rubric generation but only one rubric config can be used for a rubric-based metric. If more than one rubric config is provided, the evaluation metric must specify a rubric group key. Note that if a generation spec is specified on both a rubric config and an evaluation metric, the rubrics generated for the metric will be used for evaluation. - `predefinedRubricGenerationSpec` GoogleCloudAiplatformV1EvaluationRunMetricPredefinedMetricSpec: Specification for a pre-defined metric. - `metricSpecName` string: Required. The name of a pre-defined metric, such as "instruction_following_v1" or "text_quality_v1". - `parameters` object: Optional. The parameters needed to run the pre-defined metric. - `rubricGenerationSpec` GoogleCloudAiplatformV1EvaluationRunMetricRubricGenerationSpec: Specification for how rubrics should be generated. - `metricResourceName` string: Optional. Resource name of the metric definition. - `modelConfig` GoogleCloudAiplatformV1EvaluationRunEvaluationConfigAutoraterConfig: The autorater config used for the evaluation run. - `autoraterModel` string: Optional. The fully qualified name of the publisher model or tuned autorater endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}` - `generationConfig` GoogleCloudAiplatformV1GenerationConfig: Configuration for content generation. This message contains all the parameters that control how the model generates content. It allows you to influence the randomness, length, and structure of the output. - `audioTimestamp` boolean: Optional. If enabled, audio timestamps will be included in the request to the model. This can be useful for synchronizing audio with other modalities in the response. - `audioTranscriptionConfig` GoogleCloudAiplatformV1AudioTranscriptionConfig: Configuration for speech recognition (transcription). - … - `candidateCount` integer: Optional. The number of candidate responses to generate. A higher `candidate_count` can provide more options to choose from, but it also consumes more resources. This can be useful for generating a variety of responses and selecting the best one. - `enableAffectiveDialog` boolean: Optional. If enabled, the model will detect emotions and adapt its responses accordingly. For example, if the model detects that the user is frustrated, it may provide a more empathetic response. - `frequencyPenalty` number, float: Optional. Penalizes tokens based on their frequency in the generated text. A positive value helps to reduce the repetition of words and phrases. Valid values can range from [-2.0, 2.0]. - `imageConfig` GoogleCloudAiplatformV1ImageConfig: Configuration for image generation. This message allows you to control various aspects of image generation, such as the output format, aspect ratio, and whether the model can generate images of people. - … - `logprobs` integer: Optional. The number of top log probabilities to return for each token. This can be used to see which other tokens were considered likely candidates for a given position. A higher value will return more options, but it will also increase the size of the response. - `maxOutputTokens` integer: Optional. The maximum number of tokens to generate in the response. A token is approximately four characters. The default value varies by model. This parameter can be used to control the length of the generated text and prevent overly long responses. - `mediaResolution` 'MEDIA_RESOLUTION_UNSPECIFIED' | 'MEDIA_RESOLUTION_LOW' | 'MEDIA_RESOLUTION_MEDIUM' | 'MEDIA_RESOLUTION_HIGH': Optional. The token resolution at which input media content is sampled. This is used to control the trade-off between the quality of the response and the number of tokens used to represent the media. A higher resolution allows the model to perceive more detail, which can lead to a more nuanced response, but it will also use more tokens. This does not affect the image dimensions sent to the model. - `presencePenalty` number, float: Optional. Penalizes tokens that have already appeared in the generated text. A positive value encourages the model to generate more diverse and less repetitive text. Valid values can range from [-2.0, 2.0]. - `responseFormat` GoogleCloudAiplatformV1ResponseFormat[]: Optional. New response format field for the model to configure output formatting and delivery. - … - `responseJsonSchema` unknown - `responseLogprobs` boolean: Optional. If set to true, the log probabilities of the output tokens are returned. Log probabilities are the logarithm of the probability of a token appearing in the output. A higher log probability means the token is more likely to be generated. This can be useful for analyzing the model's confidence in its own output and for debugging. - `responseMimeType` string: Optional. The IANA standard MIME type of the response. The model will generate output that conforms to this MIME type. Supported values include 'text/plain' (default) and 'application/json'. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. Deprecated: Use `response_format` instead. - `responseModalities` string[]: Optional. The modalities of the response. The model will generate a response that includes all the specified modalities. For example, if this is set to `[TEXT, IMAGE]`, the response will include both text and an image. - `responseSchema` GoogleCloudAiplatformV1Schema: Defines the schema of input and output data. This is a subset of the [OpenAPI 3.0 Schema Object](https://spec.openapis.org/oas/v3.0.3#schema-object). - … - `routingConfig` GoogleCloudAiplatformV1GenerationConfigRoutingConfig: The configuration for routing the request to a specific model. This can be used to control which model is used for the generation, either automatically or by specifying a model name. - … - `seed` integer: Optional. A seed for the random number generator. By setting a seed, you can make the model's output mostly deterministic. For a given prompt and parameters (like temperature, top_p, etc.), the model will produce the same response every time. However, it's not a guaranteed absolute deterministic behavior. This is different from parameters like `temperature`, which control the *level* of randomness. `seed` ensures that the "random" choices the model makes are the same on every run, making it essential for testing and ensuring reproducible results. - `speechConfig` GoogleCloudAiplatformV1SpeechConfig: Configuration for speech generation. - … - `stopSequences` string[]: Optional. A list of character sequences that will stop the model from generating further tokens. If a stop sequence is generated, the output will end at that point. This is useful for controlling the length and structure of the output. For example, you can use ["\n", "###"] to stop generation at a new line or a specific marker. - `temperature` number, float: Optional. Controls the randomness of the output. A higher temperature results in more creative and diverse responses, while a lower temperature makes the output more predictable and focused. The valid range is (0.0, 2.0]. - `thinkingConfig` GoogleCloudAiplatformV1GenerationConfigThinkingConfig: Configuration for the model's thinking features. "Thinking" is a process where the model breaks down a complex task into smaller, manageable steps. This allows the model to reason about the task, plan its approach, and execute the plan to generate a high-quality response. - … - `topK` number, float: Optional. Specifies the top-k sampling threshold. The model considers only the top k most probable tokens for the next token. This can be useful for generating more coherent and less random text. For example, a `top_k` of 40 means the model will choose the next word from the 40 most likely words. - `topP` number, float: Optional. Specifies the nucleus sampling threshold. The model considers only the smallest set of tokens whose cumulative probability is at least `top_p`. This helps generate more diverse and less repetitive responses. For example, a `top_p` of 0.9 means the model considers tokens until the cumulative probability of the tokens to select from reaches 0.9. It's recommended to adjust either temperature or `top_p`, but not both. - `translationConfig` GoogleCloudAiplatformV1TranslationConfig: Config for translation features. - … - `sampleCount` integer: Optional. Number of samples for each instance in the dataset. If not specified, the default is 4. Minimum value is 1, maximum value is 32. - `promptTemplate` string: Optional. Template for the prompt used to generate rubrics. The details should be updated based on the most-recent recipe requirements. - `rubricContentType` 'RUBRIC_CONTENT_TYPE_UNSPECIFIED' | 'PROPERTY' | 'NL_QUESTION_ANSWER' | 'PYTHON_CODE_ASSERTION': Optional. The type of rubric content to be generated. - `rubricTypeOntology` string[]: Optional. An optional, pre-defined list of allowed types for generated rubrics. If this field is provided, it implies `include_rubric_type` should be true, and the generated rubric types should be chosen from this ontology. - `rubricGroupKey` string: Required. The key used to save the generated rubrics. If a generation spec is provided, this key will be used for the name of the generated rubric group. Otherwise, this key will be used to look up the existing rubric group on the evaluation item. Note that if a rubric group key is specified on both a rubric config and an evaluation metric, the key from the metric will be used to select the rubrics for evaluation. - `evaluationResults` GoogleCloudAiplatformV1EvaluationResults: The results of the evaluation run. - `evaluationSet` string: The evaluation set where item level results are stored. - `summaryMetrics` GoogleCloudAiplatformV1SummaryMetrics: The summary metrics for the evaluation run. - `failedItems` integer: Optional. The number of items that failed to be evaluated. - `metrics` object: Optional. Map of metric name to metric value. - `totalItems` integer: Optional. The total number of items that were evaluated. - `evaluationSetSnapshot` string: Output only. The specific evaluation set of the evaluation run. For runs with an evaluation set input, this will be that same set. For runs with BigQuery input, it's the sampled BigQuery dataset. - `inferenceConfigs` object: Optional. The candidate to inference config map for the evaluation run. The candidate can be up to 128 characters long and can consist of any UTF-8 characters. - `labels` object: Optional. Labels for the evaluation run. - `metadata` unknown - `name` string: Identifier. The resource name of the EvaluationRun. This is a unique identifier. Format: `projects/{project}/locations/{location}/evaluationRuns/{evaluation_run}` - `serviceAccount` string: Optional. The service account to run the evaluation run as. If set, the run uses this service account to access Google Cloud resources. Access is limited to the IAM permissions that you grant to the service account. Must be provided if an inference config sets `harbor_run_config`, `data_agent_config`. If no inference config requires this field, setting it returns an error. The caller must have the `iam.serviceAccounts.actAs` permission on this service account, and the project's [Vertex AI Service Agent](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) must have the `roles/iam.serviceAccountTokenCreator` role on it. - `state` 'STATE_UNSPECIFIED' | 'PENDING' | 'RUNNING' | 'SUCCEEDED' | 'FAILED' | 'CANCELLED' | 'INFERENCE' | 'GENERATING_RUBRICS' | 'GENERATING_LOSS_CLUSTERS': Output only. The state of the evaluation run. --- [API](https://skmtc.dev/google/apis/aiplatform.md) - [All operations](https://skmtc.dev/google/apis/aiplatform/llms.txt) - [OpenAPI document](https://skmtc.dev/google/apis/aiplatform/revisions/3aca3b348f0c?raw)