--- title: "POST /v1/projects/{projectsId}/locations/{locationsId}:evaluateInstances" method: POST path: "/v1/projects/{projectsId}/locations/{locationsId}:evaluateInstances" tags: ["projects"] --- # POST /v1/projects/{projectsId}/locations/{locationsId}:evaluateInstances `POST /v1/projects/{projectsId}/locations/{locationsId}:evaluateInstances` ## Request - Base URL: `https://aiplatform.googleapis.com` - URL: `https://aiplatform.googleapis.com/v1/projects/{projectsId}/locations/{locationsId}:evaluateInstances` - Auth: one of: - OAuth 2 and OAuth 2 - Implicit: `https://accounts.google.com/o/oauth2/auth` - Scopes: `https://www.googleapis.com/auth/aiplatform` - Authorization code: `https://accounts.google.com/o/oauth2/auth` - Token URL: `https://accounts.google.com/o/oauth2/token` scopes: `https://www.googleapis.com/auth/aiplatform` - 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` - TypeScript: (a function with typed inputs; its response is checked against the schema) Evaluates instances based on a given metric. ## Path parameters - `projectsId` string, required - `locationsId` string, required ## Request body - GoogleCloudAiplatformV1EvaluateInstancesRequest: Request message for EvaluationService.EvaluateInstances. - `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 request location is provided elsewhere in the request, this flag must be set to true or the request will fail. - `autoraterConfig` 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. - `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. - `speaker` string: Required. The name of the speaker. This should be the same as the speaker name used in the prompt. - `voiceConfig` GoogleCloudAiplatformV1VoiceConfig: Configuration for a voice. - … - `voiceConfig` GoogleCloudAiplatformV1VoiceConfig: Configuration for a voice. - `prebuiltVoiceConfig` GoogleCloudAiplatformV1PrebuiltVoiceConfig: Configuration for a prebuilt voice. - `voiceName` string: The name of the prebuilt voice to use. - `replicatedVoiceConfig` GoogleCloudAiplatformV1ReplicatedVoiceConfig: The configuration for the replicated voice to use. - `mimeType` string: Optional. The mimetype of the voice sample. The only currently supported value is `audio/wav`. This represents 16-bit signed little-endian wav data, with a 24kHz sampling rate. `mime_type` will default to `audio/wav` if not set. - `voiceSampleAudio` string, byte: Optional. The sample of the custom voice. - `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"). - `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. - `bleuInput` GoogleCloudAiplatformV1BleuInput: Input for bleu metric. - `instances` GoogleCloudAiplatformV1BleuInstance[]: Required. Repeated bleu instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` 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. - `coherenceInput` GoogleCloudAiplatformV1CoherenceInput: Input for coherence metric. - `instance` GoogleCloudAiplatformV1CoherenceInstance: Spec for coherence instance. - `prediction` string: Required. Output of the evaluated model. - `metricSpec` GoogleCloudAiplatformV1CoherenceSpec: Spec for coherence score metric. - `version` integer: Optional. Which version to use for evaluation. - `cometInput` GoogleCloudAiplatformV1CometInput: Input for Comet metric. - `instance` GoogleCloudAiplatformV1CometInstance: Spec for Comet instance - The fields used for evaluation are dependent on the comet version. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `source` string: Optional. Source text in original language. - `metricSpec` GoogleCloudAiplatformV1CometSpec: Spec for Comet metric. - `sourceLanguage` string: Optional. Source language in BCP-47 format. - `targetLanguage` string: Optional. Target language in BCP-47 format. Covers both prediction and reference. - `version` 'COMET_VERSION_UNSPECIFIED' | 'COMET_22_SRC_REF': Required. Which version to use for evaluation. - `exactMatchInput` GoogleCloudAiplatformV1ExactMatchInput: Input for exact match metric. - `instances` GoogleCloudAiplatformV1ExactMatchInstance[]: Required. Repeated exact match instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1ExactMatchSpec: Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0. - `fluencyInput` GoogleCloudAiplatformV1FluencyInput: Input for fluency metric. - `instance` GoogleCloudAiplatformV1FluencyInstance: Spec for fluency instance. - `prediction` string: Required. Output of the evaluated model. - `metricSpec` GoogleCloudAiplatformV1FluencySpec: Spec for fluency score metric. - `version` integer: Optional. Which version to use for evaluation. - `fulfillmentInput` GoogleCloudAiplatformV1FulfillmentInput: Input for fulfillment metric. - `instance` GoogleCloudAiplatformV1FulfillmentInstance: Spec for fulfillment instance. - `instruction` string: Required. Inference instruction prompt to compare prediction with. - `prediction` string: Required. Output of the evaluated model. - `metricSpec` GoogleCloudAiplatformV1FulfillmentSpec: Spec for fulfillment metric. - `version` integer: Optional. Which version to use for evaluation. - `groundednessInput` GoogleCloudAiplatformV1GroundednessInput: Input for groundedness metric. - `instance` GoogleCloudAiplatformV1GroundednessInstance: Spec for groundedness instance. - `context` string: Required. Background information provided in context used to compare against the prediction. - `prediction` string: Required. Output of the evaluated model. - `metricSpec` GoogleCloudAiplatformV1GroundednessSpec: Spec for groundedness metric. - `version` integer: Optional. Which version to use for evaluation. - `instance` GoogleCloudAiplatformV1EvaluationInstance: A single instance to be evaluated. Instances are used to specify the input data for evaluation, from simple string comparisons to complex, multi-turn model evaluations - `agentData` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentData: Deprecated: Use `agent_eval_data` instead. Contains data specific to agent evaluations. - `agentConfig` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentConfig: Deprecated: Use `google.cloud.aiplatform.master.AgentConfig` in `agent_eval_data` instead. Configuration for an Agent. - `agentId` string: Optional. Unique identifier of the agent. This ID is used to refer to this agent, e.g., in AgentEvent.author, or in the `sub_agents` field. It must be unique within the `agents` map. - `agentType` string: Optional. The type or class of the agent (e.g., "LlmAgent", "RouterAgent", "ToolUseAgent"). Useful for the autorater to understand the expected behavior of the agent. - `description` string: Optional. A high-level description of the agent's role and responsibilities. Critical for evaluating if the agent is routing tasks correctly. - `developerInstruction` GoogleCloudAiplatformV1EvaluationInstanceInstanceData: Instance data used to populate placeholders in a metric prompt template. - `contents` GoogleCloudAiplatformV1EvaluationInstanceInstanceDataContents: List of standard Content messages from Gemini API. - `contents` GoogleCloudAiplatformV1Content[]: Optional. Repeated contents. - … - `text` string: Text data. - `subAgents` string[]: Optional. The list of valid agent IDs (names) that this agent can delegate to. This defines the directed edges in the agent system graph topology. - `tools` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentConfigTools: Represents a list of tools for an agent. - `tool` GoogleCloudAiplatformV1Tool[]: Optional. List of tools: each tool can have multiple function declarations. - `codeExecution` GoogleCloudAiplatformV1ToolCodeExecution: Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. - `computerUse` GoogleCloudAiplatformV1ToolComputerUse: A tool that enables the model to interact directly with a computer environment. - … - `enterpriseWebSearch` GoogleCloudAiplatformV1EnterpriseWebSearch: Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. - … - `exaAiSearch` GoogleCloudAiplatformV1ToolExaAiSearch: ExaAiSearch tool type. A tool that uses the Exa.ai search engine for grounding. - … - `functionDeclarations` GoogleCloudAiplatformV1FunctionDeclaration[]: Optional. Function tool type. One or more function declarations to be passed to the model along with the current user query. Model may decide to call a subset of these functions by populating FunctionCall in the response. User should provide a FunctionResponse for each function call in the next turn. Based on the function responses, Model will generate the final response back to the user. Maximum 512 function declarations can be provided. - … - `googleMaps` GoogleCloudAiplatformV1GoogleMaps: Tool to retrieve public maps data for grounding, powered by Google. - … - `googleSearch` GoogleCloudAiplatformV1ToolGoogleSearch: GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google. - … - `googleSearchRetrieval` GoogleCloudAiplatformV1GoogleSearchRetrieval: Tool to retrieve public web data for grounding, powered by Google. - … - `parallelAiSearch` GoogleCloudAiplatformV1ToolParallelAiSearch: ParallelAiSearch tool type. A tool that uses the Parallel.ai search engine for grounding. - … - `retrieval` GoogleCloudAiplatformV1Retrieval: Defines a retrieval tool that model can call to access external knowledge. - … - `urlContext` GoogleCloudAiplatformV1UrlContext: Tool to support URL context. - `toolsText` string: A JSON string containing a list of tools available to an agent with info such as name, description, parameters and required parameters. - `agents` object: Optional. The static Agent Configuration. This map defines the graph structure of the agent system. Key: agent_id (matches the `author` field in events). Value: The static configuration of the agent (tools, instructions, sub-agents). - `developerInstruction` GoogleCloudAiplatformV1EvaluationInstanceInstanceData: Instance data used to populate placeholders in a metric prompt template. - `contents` GoogleCloudAiplatformV1EvaluationInstanceInstanceDataContents: List of standard Content messages from Gemini API. - `contents` GoogleCloudAiplatformV1Content[]: Optional. Repeated contents. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `text` string: Text data. - `events` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentDataEvents: Represents a list of events for an agent. - `event` GoogleCloudAiplatformV1Content[]: Optional. A list of events. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - `audioTranscription` GoogleCloudAiplatformV1AudioTranscription: The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. - … - `codeExecutionResult` GoogleCloudAiplatformV1CodeExecutionResult: Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. - … - `executableCode` GoogleCloudAiplatformV1ExecutableCode: Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the `CodeExecution` tool, in which the code will be automatically executed, and a corresponding CodeExecutionResult will also be generated. - … - `fileData` GoogleCloudAiplatformV1FileData: URI-based data. A FileData message contains a URI pointing to data of a specific media type. It is used to represent images, audio, and video stored in Google Cloud Storage. - … - `functionCall` GoogleCloudAiplatformV1FunctionCall: A predicted FunctionCall returned from the model that contains a string representing the FunctionDeclaration.name and a structured JSON object containing the parameters and their values. - … - `functionResponse` GoogleCloudAiplatformV1FunctionResponse: The result output from a FunctionCall that contains a string representing the FunctionDeclaration.name and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a `FunctionCall` made based on model prediction. - … - `inlineData` GoogleCloudAiplatformV1Blob: A content blob. A Blob contains data of a specific media type. It is used to represent images, audio, and video. - … - `mediaResolution` GoogleCloudAiplatformV1PartMediaResolution: per part media resolution. Media resolution for the input media. - … - `speechMetadata` GoogleCloudAiplatformV1SpeechMetadata: Structured Metadata Sub-Message for Part - … - `text` string: Optional. The text content of the part. When sent from the VSCode Gemini Code Assist extension, references to @mentioned items will be converted to markdown boldface text. For example `@my-repo` will be converted to and sent as `**my-repo**` by the IDE agent. - `thought` boolean: Optional. Indicates whether the `part` represents the model's thought process or reasoning. - `thoughtSignature` string, byte: Optional. An opaque signature for the thought so it can be reused in subsequent requests. - `videoMetadata` GoogleCloudAiplatformV1VideoMetadata: Provides metadata for a video, including the start and end offsets for clipping and the frame rate. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `tools` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentDataTools: Deprecated: Use `agent_eval_data` instead. Represents a list of tools for an agent. - `tool` GoogleCloudAiplatformV1Tool[]: Optional. List of tools: each tool can have multiple function declarations. - `codeExecution` GoogleCloudAiplatformV1ToolCodeExecution: Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. - `computerUse` GoogleCloudAiplatformV1ToolComputerUse: A tool that enables the model to interact directly with a computer environment. - `disabledSafetyPolicies` string[]: Optional. A list of safety policies to disable for the computer use tool. - `enablePromptInjectionDetection` boolean: Optional. Whether to enable the prompt injection detection check on the computer use request. - `environment` 'ENVIRONMENT_UNSPECIFIED' | 'ENVIRONMENT_BROWSER' | 'ENVIRONMENT_MOBILE' | 'ENVIRONMENT_DESKTOP': Required. The target environment where the computer use tool operates. - `excludedPredefinedFunctions` string[]: Optional. A list of predefined functions to explicitly exclude from the model call. By default, [predefined functions](https://cloud.google.com/vertex-ai/generative-ai/docs/computer-use#supported-actions) are included. Excluding functions allows for a more restricted action space or custom definitions for predefined functions. - `enterpriseWebSearch` GoogleCloudAiplatformV1EnterpriseWebSearch: Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. - `blockingConfidence` 'PHISH_BLOCK_THRESHOLD_UNSPECIFIED' | 'BLOCK_LOW_AND_ABOVE' | 'BLOCK_MEDIUM_AND_ABOVE' | 'BLOCK_HIGH_AND_ABOVE' | 'BLOCK_HIGHER_AND_ABOVE' | 'BLOCK_VERY_HIGH_AND_ABOVE' | 'BLOCK_ONLY_EXTREMELY_HIGH': Optional. Sites with confidence level chosen & above this value will be blocked from the search results. - `excludeDomains` string[]: Optional. List of domains to be excluded from the search results. The default limit is 2000 domains. - `exaAiSearch` GoogleCloudAiplatformV1ToolExaAiSearch: ExaAiSearch tool type. A tool that uses the Exa.ai search engine for grounding. - `apiKey` string: Required. The API key for ExaAiSearch. - `customConfigs` object: Optional. This field can be used to pass any parameter from the Exa.ai Search API. - `functionDeclarations` GoogleCloudAiplatformV1FunctionDeclaration[]: Optional. Function tool type. One or more function declarations to be passed to the model along with the current user query. Model may decide to call a subset of these functions by populating FunctionCall in the response. User should provide a FunctionResponse for each function call in the next turn. Based on the function responses, Model will generate the final response back to the user. Maximum 512 function declarations can be provided. - `behavior` 'UNSPECIFIED' | 'BLOCKING' | 'NON_BLOCKING': Optional. Specifies the function Behavior. If not specified, the system keeps the current function call behavior. This field is currently only supported by the BidiGenerateContent method. - `description` string: Optional. Description and purpose of the function. Model uses it to decide how and whether to call the function. - `name` string: Required. The name of the function to call. Must start with a letter or an underscore. Must be a-z, A-Z, 0-9, or contain underscores, dots, colons and dashes, with a maximum length of 128. - `parameters` 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). - … - `parametersJsonSchema` unknown - `response` 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). - … - `responseJsonSchema` unknown - `googleMaps` GoogleCloudAiplatformV1GoogleMaps: Tool to retrieve public maps data for grounding, powered by Google. - `enableWidget` boolean: Optional. Deprecated: The Google Maps contextual widget behavior in Grounding with Google Maps is being deprecated; this field is planned for removal and no longer has any effect once removed. If true, include the widget context token in the response. - `groundingTypes` GoogleCloudAiplatformV1GoogleMapsGroundingTypes: Defines the types of Google Maps grounding that can be enabled and their configurations. - … - `googleSearch` GoogleCloudAiplatformV1ToolGoogleSearch: GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google. - `blockingConfidence` 'PHISH_BLOCK_THRESHOLD_UNSPECIFIED' | 'BLOCK_LOW_AND_ABOVE' | 'BLOCK_MEDIUM_AND_ABOVE' | 'BLOCK_HIGH_AND_ABOVE' | 'BLOCK_HIGHER_AND_ABOVE' | 'BLOCK_VERY_HIGH_AND_ABOVE' | 'BLOCK_ONLY_EXTREMELY_HIGH': Optional. Sites with confidence level chosen & above this value will be blocked from the search results. - `excludeDomains` string[]: Optional. List of domains to be excluded from the search results. The default limit is 2000 domains. Example: ["amazon.com", "facebook.com"]. - `searchTypes` GoogleCloudAiplatformV1ToolGoogleSearchSearchTypes: Different types of search that can be enabled on the GoogleSearch tool. - … - `googleSearchRetrieval` GoogleCloudAiplatformV1GoogleSearchRetrieval: Tool to retrieve public web data for grounding, powered by Google. - `dynamicRetrievalConfig` GoogleCloudAiplatformV1DynamicRetrievalConfig: Describes the options to customize dynamic retrieval. - … - `parallelAiSearch` GoogleCloudAiplatformV1ToolParallelAiSearch: ParallelAiSearch tool type. A tool that uses the Parallel.ai search engine for grounding. - `apiKey` string: Optional. The API key for ParallelAiSearch. If an API key is not provided, the system will attempt to verify access by checking for an active Parallel.ai subscription through the Google Cloud Marketplace. See https://docs.parallel.ai/search/search-quickstart for more details. - `customConfigs` object: Optional. Custom configs for ParallelAiSearch. This field can be used to pass any parameter from the Parallel.ai Search API. See the Parallel.ai documentation for the full list of available parameters and their usage: https://docs.parallel.ai/api-reference/search-beta/search Currently only `source_policy`, `excerpts`, `max_results`, `mode`, `fetch_policy` can be set via this field. For example: { "source_policy": { "include_domains": ["google.com", "wikipedia.org"], "exclude_domains": ["example.com"] }, "fetch_policy": { "max_age_seconds": 3600 } } - `enableDataRetention` boolean: Optional. Deprecated: Use `enable_zero_data_retention` instead. Instructs Vertex Grounding to use Parallel's Zero Data Retention Marketplace product. If this value is "false" or omitted, the Parallel Web Search for Grounding standard subscription will be used. If this value is "true", the Parallel Web Search for Grounding - ZDR subscription will be used. - `enableZeroDataRetention` boolean: Optional. Instructs Vertex Grounding to use Parallel's Zero Data Retention Marketplace product. If this value is "false" or omitted, the Parallel Web Search for Grounding standard subscription will be used. If this value is "true", the Parallel Web Search for Grounding - ZDR subscription will be used. - `retrieval` GoogleCloudAiplatformV1Retrieval: Defines a retrieval tool that model can call to access external knowledge. - `disableAttribution` boolean: Optional. Deprecated. This option is no longer supported. - `externalApi` GoogleCloudAiplatformV1ExternalApi: Retrieve from data source powered by external API for grounding. The external API is not owned by Google, but need to follow the pre-defined API spec. - … - `vertexAiSearch` GoogleCloudAiplatformV1VertexAISearch: Retrieve from Vertex AI Search datastore or engine for grounding. datastore and engine are mutually exclusive. See https://cloud.google.com/products/agent-builder - … - `vertexRagStore` GoogleCloudAiplatformV1VertexRagStore: Retrieve from Vertex RAG Store for grounding. - … - `urlContext` GoogleCloudAiplatformV1UrlContext: Tool to support URL context. - `toolsText` string: A JSON string containing a list of tools available to an agent with info such as name, description, parameters and required parameters. - `turns` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentDataConversationTurn[]: Optional. The chronological list of conversation turns. Each turn represents a logical execution cycle (e.g., User Input -> Agent Response). - `events` GoogleCloudAiplatformV1EvaluationInstanceDeprecatedAgentDataAgentEvent[]: Optional. The list of events that occurred during this turn. - `activeTools` GoogleCloudAiplatformV1Tool[]: Optional. The list of tools that were active/available to the agent at the time of this event. This overrides the `AgentConfig.tools` if set. - `codeExecution` GoogleCloudAiplatformV1ToolCodeExecution: Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. - `computerUse` GoogleCloudAiplatformV1ToolComputerUse: A tool that enables the model to interact directly with a computer environment. - … - `enterpriseWebSearch` GoogleCloudAiplatformV1EnterpriseWebSearch: Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. - … - `exaAiSearch` GoogleCloudAiplatformV1ToolExaAiSearch: ExaAiSearch tool type. A tool that uses the Exa.ai search engine for grounding. - … - `functionDeclarations` GoogleCloudAiplatformV1FunctionDeclaration[]: Optional. Function tool type. One or more function declarations to be passed to the model along with the current user query. Model may decide to call a subset of these functions by populating FunctionCall in the response. User should provide a FunctionResponse for each function call in the next turn. Based on the function responses, Model will generate the final response back to the user. Maximum 512 function declarations can be provided. - … - `googleMaps` GoogleCloudAiplatformV1GoogleMaps: Tool to retrieve public maps data for grounding, powered by Google. - … - `googleSearch` GoogleCloudAiplatformV1ToolGoogleSearch: GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google. - … - `googleSearchRetrieval` GoogleCloudAiplatformV1GoogleSearchRetrieval: Tool to retrieve public web data for grounding, powered by Google. - … - `parallelAiSearch` GoogleCloudAiplatformV1ToolParallelAiSearch: ParallelAiSearch tool type. A tool that uses the Parallel.ai search engine for grounding. - … - `retrieval` GoogleCloudAiplatformV1Retrieval: Defines a retrieval tool that model can call to access external knowledge. - … - `urlContext` GoogleCloudAiplatformV1UrlContext: Tool to support URL context. - `author` string: Required. The ID of the agent or entity that generated this event. - `content` GoogleCloudAiplatformV1Content: The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `eventTime` string, google-datetime: Optional. The timestamp when the event occurred. - `stateDelta` object: Optional. The change in the session state caused by this event. This is a key-value map of fields that were modified or added by the event. - `turnId` string: Optional. A unique identifier for the turn. Useful for referencing specific turns across systems. - `turnIndex` integer: Required. The 0-based index of the turn in the conversation sequence. - `agentEvalData` GoogleCloudAiplatformV1AgentData: Represents data specific to multi-turn agent evaluations. - `agents` object: Optional. A map containing the static configurations for each agent in the system. Key: agent_id (matches the `author` field in events). Value: The static configuration of the agent. - `turns` GoogleCloudAiplatformV1ConversationTurn[]: Optional. A chronological list of conversation turns. Each turn represents a logical execution cycle (e.g., User Input -> Agent Response). - `events` GoogleCloudAiplatformV1AgentEvent[]: Optional. The list of events that occurred during this turn. - `activeTools` GoogleCloudAiplatformV1Tool[]: Optional. The list of tools that were active/available to the agent at the time of this event. This overrides the `AgentConfig.tools` if set. - `codeExecution` GoogleCloudAiplatformV1ToolCodeExecution: Tool that executes code generated by the model, and automatically returns the result to the model. See also ExecutableCode and CodeExecutionResult, which are input and output to this tool. - `computerUse` GoogleCloudAiplatformV1ToolComputerUse: A tool that enables the model to interact directly with a computer environment. - … - `enterpriseWebSearch` GoogleCloudAiplatformV1EnterpriseWebSearch: Tool to search public web data, powered by Vertex AI Search and Sec4 compliance. - … - `exaAiSearch` GoogleCloudAiplatformV1ToolExaAiSearch: ExaAiSearch tool type. A tool that uses the Exa.ai search engine for grounding. - … - `functionDeclarations` GoogleCloudAiplatformV1FunctionDeclaration[]: Optional. Function tool type. One or more function declarations to be passed to the model along with the current user query. Model may decide to call a subset of these functions by populating FunctionCall in the response. User should provide a FunctionResponse for each function call in the next turn. Based on the function responses, Model will generate the final response back to the user. Maximum 512 function declarations can be provided. - … - `googleMaps` GoogleCloudAiplatformV1GoogleMaps: Tool to retrieve public maps data for grounding, powered by Google. - … - `googleSearch` GoogleCloudAiplatformV1ToolGoogleSearch: GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google. - … - `googleSearchRetrieval` GoogleCloudAiplatformV1GoogleSearchRetrieval: Tool to retrieve public web data for grounding, powered by Google. - … - `parallelAiSearch` GoogleCloudAiplatformV1ToolParallelAiSearch: ParallelAiSearch tool type. A tool that uses the Parallel.ai search engine for grounding. - … - `retrieval` GoogleCloudAiplatformV1Retrieval: Defines a retrieval tool that model can call to access external knowledge. - … - `urlContext` GoogleCloudAiplatformV1UrlContext: Tool to support URL context. - `author` string: Required. The ID of the agent or entity that generated this event. Use "user" to denote events generated by the end-user. - `content` GoogleCloudAiplatformV1Content: The structured data content of a message. A Content message contains a `role` field, which indicates the producer of the content, and a `parts` field, which contains the multi-part data of the message. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `eventTime` string, google-datetime: Optional. The timestamp when the event occurred. - `stateDelta` object: Optional. The change in the session state caused by this event. This is a key-value map of fields that were modified or added by the event. - `turnId` string: Optional. A unique identifier for the turn. Useful for referencing specific turns across systems. - `turnIndex` integer: Required. The 0-based index of the turn in the conversation sequence. - `interactionsDataSource` GoogleCloudAiplatformV1EvaluationInstanceInteractionsDataSource: Source for populating `AgentData` from an Interactions API interaction. - `geminiAgentConfig` GoogleCloudAiplatformV1GeminiAgentConfig: Config for scraping a Gemini Agent (a Vertex AI Agent resource scraped via the Vertex Interactions API). - `geminiAgent` string: Required. The resource name of the Gemini Agent. Format: `projects/{project}/locations/{location}/agents/{agent}`. For example: `projects/123/locations/us-central1/agents/my-agent`. - `interaction` string: Required. The interaction to evaluate. Format: `projects/{project}/locations/{location}/interactions/{interaction}`. - `otherData` GoogleCloudAiplatformV1EvaluationInstanceMapInstance: Instance data specified as a map. - `mapInstance` object: Optional. Map of instance data. - `prompt` GoogleCloudAiplatformV1EvaluationInstanceInstanceData: Instance data used to populate placeholders in a metric prompt template. - `contents` GoogleCloudAiplatformV1EvaluationInstanceInstanceDataContents: List of standard Content messages from Gemini API. - `contents` GoogleCloudAiplatformV1Content[]: Optional. Repeated contents. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - `audioTranscription` GoogleCloudAiplatformV1AudioTranscription: The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. - … - `codeExecutionResult` GoogleCloudAiplatformV1CodeExecutionResult: Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. - … - `executableCode` GoogleCloudAiplatformV1ExecutableCode: Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the `CodeExecution` tool, in which the code will be automatically executed, and a corresponding CodeExecutionResult will also be generated. - … - `fileData` GoogleCloudAiplatformV1FileData: URI-based data. A FileData message contains a URI pointing to data of a specific media type. It is used to represent images, audio, and video stored in Google Cloud Storage. - … - `functionCall` GoogleCloudAiplatformV1FunctionCall: A predicted FunctionCall returned from the model that contains a string representing the FunctionDeclaration.name and a structured JSON object containing the parameters and their values. - … - `functionResponse` GoogleCloudAiplatformV1FunctionResponse: The result output from a FunctionCall that contains a string representing the FunctionDeclaration.name and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a `FunctionCall` made based on model prediction. - … - `inlineData` GoogleCloudAiplatformV1Blob: A content blob. A Blob contains data of a specific media type. It is used to represent images, audio, and video. - … - `mediaResolution` GoogleCloudAiplatformV1PartMediaResolution: per part media resolution. Media resolution for the input media. - … - `speechMetadata` GoogleCloudAiplatformV1SpeechMetadata: Structured Metadata Sub-Message for Part - … - `text` string: Optional. The text content of the part. When sent from the VSCode Gemini Code Assist extension, references to @mentioned items will be converted to markdown boldface text. For example `@my-repo` will be converted to and sent as `**my-repo**` by the IDE agent. - `thought` boolean: Optional. Indicates whether the `part` represents the model's thought process or reasoning. - `thoughtSignature` string, byte: Optional. An opaque signature for the thought so it can be reused in subsequent requests. - `videoMetadata` GoogleCloudAiplatformV1VideoMetadata: Provides metadata for a video, including the start and end offsets for clipping and the frame rate. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `text` string: Text data. - `reference` GoogleCloudAiplatformV1EvaluationInstanceInstanceData: Instance data used to populate placeholders in a metric prompt template. - `contents` GoogleCloudAiplatformV1EvaluationInstanceInstanceDataContents: List of standard Content messages from Gemini API. - `contents` GoogleCloudAiplatformV1Content[]: Optional. Repeated contents. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - `audioTranscription` GoogleCloudAiplatformV1AudioTranscription: The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. - … - `codeExecutionResult` GoogleCloudAiplatformV1CodeExecutionResult: Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. - … - `executableCode` GoogleCloudAiplatformV1ExecutableCode: Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the `CodeExecution` tool, in which the code will be automatically executed, and a corresponding CodeExecutionResult will also be generated. - … - `fileData` GoogleCloudAiplatformV1FileData: URI-based data. A FileData message contains a URI pointing to data of a specific media type. It is used to represent images, audio, and video stored in Google Cloud Storage. - … - `functionCall` GoogleCloudAiplatformV1FunctionCall: A predicted FunctionCall returned from the model that contains a string representing the FunctionDeclaration.name and a structured JSON object containing the parameters and their values. - … - `functionResponse` GoogleCloudAiplatformV1FunctionResponse: The result output from a FunctionCall that contains a string representing the FunctionDeclaration.name and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a `FunctionCall` made based on model prediction. - … - `inlineData` GoogleCloudAiplatformV1Blob: A content blob. A Blob contains data of a specific media type. It is used to represent images, audio, and video. - … - `mediaResolution` GoogleCloudAiplatformV1PartMediaResolution: per part media resolution. Media resolution for the input media. - … - `speechMetadata` GoogleCloudAiplatformV1SpeechMetadata: Structured Metadata Sub-Message for Part - … - `text` string: Optional. The text content of the part. When sent from the VSCode Gemini Code Assist extension, references to @mentioned items will be converted to markdown boldface text. For example `@my-repo` will be converted to and sent as `**my-repo**` by the IDE agent. - `thought` boolean: Optional. Indicates whether the `part` represents the model's thought process or reasoning. - `thoughtSignature` string, byte: Optional. An opaque signature for the thought so it can be reused in subsequent requests. - `videoMetadata` GoogleCloudAiplatformV1VideoMetadata: Provides metadata for a video, including the start and end offsets for clipping and the frame rate. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `text` string: Text data. - `response` GoogleCloudAiplatformV1EvaluationInstanceInstanceData: Instance data used to populate placeholders in a metric prompt template. - `contents` GoogleCloudAiplatformV1EvaluationInstanceInstanceDataContents: List of standard Content messages from Gemini API. - `contents` GoogleCloudAiplatformV1Content[]: Optional. Repeated contents. - `parts` GoogleCloudAiplatformV1Part[]: Required. A list of Part objects that make up a single message. Parts of a message can have different MIME types. A Content message must have at least one Part. - `audioTranscription` GoogleCloudAiplatformV1AudioTranscription: The transcription of an audio part. For multi-speaker audio, each speaker segment is a separate `Part` with its own `AudioTranscription` carrying the `speaker_label`. - … - `codeExecutionResult` GoogleCloudAiplatformV1CodeExecutionResult: Result of executing the ExecutableCode. Generated only when the `CodeExecution` tool is used. - … - `executableCode` GoogleCloudAiplatformV1ExecutableCode: Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the `CodeExecution` tool, in which the code will be automatically executed, and a corresponding CodeExecutionResult will also be generated. - … - `fileData` GoogleCloudAiplatformV1FileData: URI-based data. A FileData message contains a URI pointing to data of a specific media type. It is used to represent images, audio, and video stored in Google Cloud Storage. - … - `functionCall` GoogleCloudAiplatformV1FunctionCall: A predicted FunctionCall returned from the model that contains a string representing the FunctionDeclaration.name and a structured JSON object containing the parameters and their values. - … - `functionResponse` GoogleCloudAiplatformV1FunctionResponse: The result output from a FunctionCall that contains a string representing the FunctionDeclaration.name and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a `FunctionCall` made based on model prediction. - … - `inlineData` GoogleCloudAiplatformV1Blob: A content blob. A Blob contains data of a specific media type. It is used to represent images, audio, and video. - … - `mediaResolution` GoogleCloudAiplatformV1PartMediaResolution: per part media resolution. Media resolution for the input media. - … - `speechMetadata` GoogleCloudAiplatformV1SpeechMetadata: Structured Metadata Sub-Message for Part - … - `text` string: Optional. The text content of the part. When sent from the VSCode Gemini Code Assist extension, references to @mentioned items will be converted to markdown boldface text. For example `@my-repo` will be converted to and sent as `**my-repo**` by the IDE agent. - `thought` boolean: Optional. Indicates whether the `part` represents the model's thought process or reasoning. - `thoughtSignature` string, byte: Optional. An opaque signature for the thought so it can be reused in subsequent requests. - `videoMetadata` GoogleCloudAiplatformV1VideoMetadata: Provides metadata for a video, including the start and end offsets for clipping and the frame rate. - … - `role` string: Optional. The producer of the content. Must be either 'user' or 'model'. If not set, the service will default to 'user'. - `text` string: Text data. - `rubricGroups` object: Optional. Named groups of rubrics associated with the prompt. This is used for rubric-based evaluations where rubrics can be referenced by a key. The key could represent versions, associated metrics, etc. - `location` string: Required. The resource name of the Location to evaluate the instances. Format: `projects/{project}/locations/{location}` - `metricSources` GoogleCloudAiplatformV1MetricSource[]: Optional. The metrics (either inline or registered) used for evaluation. Currently, we only support evaluating a single metric. If multiple metrics are provided, only the first one will be evaluated. - `metric` 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. - `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. - … - `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. - `parsingFunction` string: Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - `rubricGenerationSpec` GoogleCloudAiplatformV1RubricGenerationSpec: Specification for how rubrics should be generated. - `modelConfig` 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. - `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. Resource name for registered metric. - `metrics` GoogleCloudAiplatformV1Metric[]: The metrics used for evaluation. Currently, we only support evaluating a single metric. If multiple metrics are provided, only the first one will be evaluated. - `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. - `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. - … - `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. - … - `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. - … - `image` GoogleCloudAiplatformV1ImageResponseFormat: Configuration for image-specific output formatting. - … - `text` GoogleCloudAiplatformV1TextResponseFormat: Configuration for text-specific output formatting. - … - `video` GoogleCloudAiplatformV1VideoResponseFormat: Configuration for video-specific output formatting. - … - `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. - … - `manualMode` GoogleCloudAiplatformV1GenerationConfigRoutingConfigManualRoutingMode: The configuration for manual routing. When manual routing is specified, the model will be selected based on the model name provided. - … - `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. - … - `voiceConfig` GoogleCloudAiplatformV1VoiceConfig: Configuration for a voice. - … - `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"). - `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. - `parsingFunction` string: Required. Python function for parsing results. The function should be defined within this string. The function takes a list of strings (LLM responses) and should return either a list of dictionaries (for rubrics) or a single dictionary (for a metric result). Example function signature: def parse(responses: list[str]) -> list[dict[str, Any]] | dict[str, Any]: When parsing rubrics, return a list of dictionaries, where each dictionary represents a Rubric. Example for rubrics: [ { "content": {"property": {"description": "The response is factual."}}, "type": "FACTUALITY", "importance": "HIGH" }, { "content": {"property": {"description": "The response is fluent."}}, "type": "FLUENCY", "importance": "MEDIUM" } ] When parsing critique results, return a dictionary representing a MetricResult. Example for a metric result: { "score": 0.8, "explanation": "The model followed most instructions.", "rubric_verdicts": [...] } ... code for result extraction and aggregation - `rubricGenerationSpec` GoogleCloudAiplatformV1RubricGenerationSpec: Specification for how rubrics should be generated. - `modelConfig` 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. - `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. - … - `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. - `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. - `metricxInput` GoogleCloudAiplatformV1MetricxInput: Input for MetricX metric. - `instance` GoogleCloudAiplatformV1MetricxInstance: Spec for MetricX instance - The fields used for evaluation are dependent on the MetricX version. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `source` string: Optional. Source text in original language. - `metricSpec` GoogleCloudAiplatformV1MetricxSpec: Spec for MetricX metric. - `sourceLanguage` string: Optional. Source language in BCP-47 format. - `targetLanguage` string: Optional. Target language in BCP-47 format. Covers both prediction and reference. - `version` 'METRICX_VERSION_UNSPECIFIED' | 'METRICX_24_REF' | 'METRICX_24_SRC' | 'METRICX_24_SRC_REF': Required. Which version to use for evaluation. - `pairwiseMetricInput` GoogleCloudAiplatformV1PairwiseMetricInput: Input for pairwise metric. - `instance` GoogleCloudAiplatformV1PairwiseMetricInstance: Pairwise metric instance. Usually one instance corresponds to one row in an evaluation dataset. - `contentMapInstance` GoogleCloudAiplatformV1ContentMap: Map of placeholder in metric prompt template to contents of model input. - `values` object: Optional. Map of placeholder to contents. - `jsonInstance` string: Instance specified as a json string. String key-value pairs are expected in the json_instance to render PairwiseMetricSpec.instance_prompt_template. - `metricSpec` 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. - `pairwiseQuestionAnsweringQualityInput` GoogleCloudAiplatformV1PairwiseQuestionAnsweringQualityInput: Input for pairwise question answering quality metric. - `instance` GoogleCloudAiplatformV1PairwiseQuestionAnsweringQualityInstance: Spec for pairwise question answering quality instance. - `baselinePrediction` string: Required. Output of the baseline model. - `context` string: Required. Text to answer the question. - `instruction` string: Required. Question Answering prompt for LLM. - `prediction` string: Required. Output of the candidate model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1PairwiseQuestionAnsweringQualitySpec: Spec for pairwise question answering quality score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute question answering quality. - `version` integer: Optional. Which version to use for evaluation. - `pairwiseSummarizationQualityInput` GoogleCloudAiplatformV1PairwiseSummarizationQualityInput: Input for pairwise summarization quality metric. - `instance` GoogleCloudAiplatformV1PairwiseSummarizationQualityInstance: Spec for pairwise summarization quality instance. - `baselinePrediction` string: Required. Output of the baseline model. - `context` string: Required. Text to be summarized. - `instruction` string: Required. Summarization prompt for LLM. - `prediction` string: Required. Output of the candidate model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1PairwiseSummarizationQualitySpec: Spec for pairwise summarization quality score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute pairwise summarization quality. - `version` integer: Optional. Which version to use for evaluation. - `pointwiseMetricInput` GoogleCloudAiplatformV1PointwiseMetricInput: Input for pointwise metric. - `instance` GoogleCloudAiplatformV1PointwiseMetricInstance: Pointwise metric instance. Usually one instance corresponds to one row in an evaluation dataset. - `contentMapInstance` GoogleCloudAiplatformV1ContentMap: Map of placeholder in metric prompt template to contents of model input. - `values` object: Optional. Map of placeholder to contents. - `jsonInstance` string: Instance specified as a json string. String key-value pairs are expected in the json_instance to render PointwiseMetricSpec.instance_prompt_template. - `metricSpec` 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. - `questionAnsweringCorrectnessInput` GoogleCloudAiplatformV1QuestionAnsweringCorrectnessInput: Input for question answering correctness metric. - `instance` GoogleCloudAiplatformV1QuestionAnsweringCorrectnessInstance: Spec for question answering correctness instance. - `context` string: Optional. Text provided as context to answer the question. - `instruction` string: Required. The question asked and other instruction in the inference prompt. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1QuestionAnsweringCorrectnessSpec: Spec for question answering correctness metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute question answering correctness. - `version` integer: Optional. Which version to use for evaluation. - `questionAnsweringHelpfulnessInput` GoogleCloudAiplatformV1QuestionAnsweringHelpfulnessInput: Input for question answering helpfulness metric. - `instance` GoogleCloudAiplatformV1QuestionAnsweringHelpfulnessInstance: Spec for question answering helpfulness instance. - `context` string: Optional. Text provided as context to answer the question. - `instruction` string: Required. The question asked and other instruction in the inference prompt. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1QuestionAnsweringHelpfulnessSpec: Spec for question answering helpfulness metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute question answering helpfulness. - `version` integer: Optional. Which version to use for evaluation. - `questionAnsweringQualityInput` GoogleCloudAiplatformV1QuestionAnsweringQualityInput: Input for question answering quality metric. - `instance` GoogleCloudAiplatformV1QuestionAnsweringQualityInstance: Spec for question answering quality instance. - `context` string: Required. Text to answer the question. - `instruction` string: Required. Question Answering prompt for LLM. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1QuestionAnsweringQualitySpec: Spec for question answering quality score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute question answering quality. - `version` integer: Optional. Which version to use for evaluation. - `questionAnsweringRelevanceInput` GoogleCloudAiplatformV1QuestionAnsweringRelevanceInput: Input for question answering relevance metric. - `instance` GoogleCloudAiplatformV1QuestionAnsweringRelevanceInstance: Spec for question answering relevance instance. - `context` string: Optional. Text provided as context to answer the question. - `instruction` string: Required. The question asked and other instruction in the inference prompt. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1QuestionAnsweringRelevanceSpec: Spec for question answering relevance metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute question answering relevance. - `version` integer: Optional. Which version to use for evaluation. - `rougeInput` GoogleCloudAiplatformV1RougeInput: Input for rouge metric. - `instances` GoogleCloudAiplatformV1RougeInstance[]: Required. Repeated rouge instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` 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. - `rubricBasedInstructionFollowingInput` GoogleCloudAiplatformV1RubricBasedInstructionFollowingInput: Instance and metric spec for RubricBasedInstructionFollowing metric. - `instance` GoogleCloudAiplatformV1RubricBasedInstructionFollowingInstance: Instance for RubricBasedInstructionFollowing metric - one instance corresponds to one row in an evaluation dataset. - `jsonInstance` string: Required. Instance specified as a json string. String key-value pairs are expected in the json_instance to render RubricBasedInstructionFollowing prompt templates. - `metricSpec` GoogleCloudAiplatformV1RubricBasedInstructionFollowingSpec: Spec for RubricBasedInstructionFollowing metric - returns rubrics and verdicts corresponding to rubrics along with overall score. - `safetyInput` GoogleCloudAiplatformV1SafetyInput: Input for safety metric. - `instance` GoogleCloudAiplatformV1SafetyInstance: Spec for safety instance. - `prediction` string: Required. Output of the evaluated model. - `metricSpec` GoogleCloudAiplatformV1SafetySpec: Spec for safety metric. - `version` integer: Optional. Which version to use for evaluation. - `summarizationHelpfulnessInput` GoogleCloudAiplatformV1SummarizationHelpfulnessInput: Input for summarization helpfulness metric. - `instance` GoogleCloudAiplatformV1SummarizationHelpfulnessInstance: Spec for summarization helpfulness instance. - `context` string: Required. Text to be summarized. - `instruction` string: Optional. Summarization prompt for LLM. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1SummarizationHelpfulnessSpec: Spec for summarization helpfulness score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute summarization helpfulness. - `version` integer: Optional. Which version to use for evaluation. - `summarizationQualityInput` GoogleCloudAiplatformV1SummarizationQualityInput: Input for summarization quality metric. - `instance` GoogleCloudAiplatformV1SummarizationQualityInstance: Spec for summarization quality instance. - `context` string: Required. Text to be summarized. - `instruction` string: Required. Summarization prompt for LLM. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1SummarizationQualitySpec: Spec for summarization quality score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute summarization quality. - `version` integer: Optional. Which version to use for evaluation. - `summarizationVerbosityInput` GoogleCloudAiplatformV1SummarizationVerbosityInput: Input for summarization verbosity metric. - `instance` GoogleCloudAiplatformV1SummarizationVerbosityInstance: Spec for summarization verbosity instance. - `context` string: Required. Text to be summarized. - `instruction` string: Optional. Summarization prompt for LLM. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Optional. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1SummarizationVerbositySpec: Spec for summarization verbosity score metric. - `useReference` boolean: Optional. Whether to use instance.reference to compute summarization verbosity. - `version` integer: Optional. Which version to use for evaluation. - `toolCallValidInput` GoogleCloudAiplatformV1ToolCallValidInput: Input for tool call valid metric. - `instances` GoogleCloudAiplatformV1ToolCallValidInstance[]: Required. Repeated tool call valid instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1ToolCallValidSpec: Spec for tool call valid metric. - `toolNameMatchInput` GoogleCloudAiplatformV1ToolNameMatchInput: Input for tool name match metric. - `instances` GoogleCloudAiplatformV1ToolNameMatchInstance[]: Required. Repeated tool name match instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1ToolNameMatchSpec: Spec for tool name match metric. - `toolParameterKeyMatchInput` GoogleCloudAiplatformV1ToolParameterKeyMatchInput: Input for tool parameter key match metric. - `instances` GoogleCloudAiplatformV1ToolParameterKeyMatchInstance[]: Required. Repeated tool parameter key match instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1ToolParameterKeyMatchSpec: Spec for tool parameter key match metric. - `toolParameterKvMatchInput` GoogleCloudAiplatformV1ToolParameterKVMatchInput: Input for tool parameter key value match metric. - `instances` GoogleCloudAiplatformV1ToolParameterKVMatchInstance[]: Required. Repeated tool parameter key value match instances. - `prediction` string: Required. Output of the evaluated model. - `reference` string: Required. Ground truth used to compare against the prediction. - `metricSpec` GoogleCloudAiplatformV1ToolParameterKVMatchSpec: Spec for tool parameter key value match metric. - `useStrictStringMatch` boolean: Optional. Whether to use STRICT string match on parameter values. - `trajectoryAnyOrderMatchInput` GoogleCloudAiplatformV1TrajectoryAnyOrderMatchInput: Instances and metric spec for TrajectoryAnyOrderMatch metric. - `instances` GoogleCloudAiplatformV1TrajectoryAnyOrderMatchInstance[]: Required. Repeated TrajectoryAnyOrderMatch instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `referenceTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectoryAnyOrderMatchSpec: Spec for TrajectoryAnyOrderMatch metric - returns 1 if all tool calls in the reference trajectory appear in the predicted trajectory in any order, else 0. - `trajectoryExactMatchInput` GoogleCloudAiplatformV1TrajectoryExactMatchInput: Instances and metric spec for TrajectoryExactMatch metric. - `instances` GoogleCloudAiplatformV1TrajectoryExactMatchInstance[]: Required. Repeated TrajectoryExactMatch instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `referenceTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectoryExactMatchSpec: Spec for TrajectoryExactMatch metric - returns 1 if tool calls in the reference trajectory exactly match the predicted trajectory, else 0. - `trajectoryInOrderMatchInput` GoogleCloudAiplatformV1TrajectoryInOrderMatchInput: Instances and metric spec for TrajectoryInOrderMatch metric. - `instances` GoogleCloudAiplatformV1TrajectoryInOrderMatchInstance[]: Required. Repeated TrajectoryInOrderMatch instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `referenceTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectoryInOrderMatchSpec: Spec for TrajectoryInOrderMatch metric - returns 1 if tool calls in the reference trajectory appear in the predicted trajectory in the same order, else 0. - `trajectoryPrecisionInput` GoogleCloudAiplatformV1TrajectoryPrecisionInput: Instances and metric spec for TrajectoryPrecision metric. - `instances` GoogleCloudAiplatformV1TrajectoryPrecisionInstance[]: Required. Repeated TrajectoryPrecision instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `referenceTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectoryPrecisionSpec: Spec for TrajectoryPrecision metric - returns a float score based on average precision of individual tool calls. - `trajectoryRecallInput` GoogleCloudAiplatformV1TrajectoryRecallInput: Instances and metric spec for TrajectoryRecall metric. - `instances` GoogleCloudAiplatformV1TrajectoryRecallInstance[]: Required. Repeated TrajectoryRecall instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `referenceTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectoryRecallSpec: Spec for TrajectoryRecall metric - returns a float score based on average recall of individual tool calls. - `trajectorySingleToolUseInput` GoogleCloudAiplatformV1TrajectorySingleToolUseInput: Instances and metric spec for TrajectorySingleToolUse metric. - `instances` GoogleCloudAiplatformV1TrajectorySingleToolUseInstance[]: Required. Repeated TrajectorySingleToolUse instance. - `predictedTrajectory` GoogleCloudAiplatformV1Trajectory: Spec for trajectory. - `toolCalls` GoogleCloudAiplatformV1ToolCall[]: Required. Tool calls in the trajectory. - `toolInput` string: Optional. Spec for tool input - `toolName` string: Required. Spec for tool name - `metricSpec` GoogleCloudAiplatformV1TrajectorySingleToolUseSpec: Spec for TrajectorySingleToolUse metric - returns 1 if tool is present in the predicted trajectory, else 0. - `toolName` string: Required. Spec for tool name to be checked for in the predicted trajectory. ## Response `200` Successful response - GoogleCloudAiplatformV1EvaluateInstancesResponse: Response message for EvaluationService.EvaluateInstances. - `bleuResults` GoogleCloudAiplatformV1BleuResults: Results for bleu metric. - `bleuMetricValues` GoogleCloudAiplatformV1BleuMetricValue[]: Output only. Bleu metric values. - `score` number, float: Output only. Bleu score. - `coherenceResult` GoogleCloudAiplatformV1CoherenceResult: Spec for coherence result. - `confidence` number, float: Output only. Confidence for coherence score. - `explanation` string: Output only. Explanation for coherence score. - `score` number, float: Output only. Coherence score. - `cometResult` GoogleCloudAiplatformV1CometResult: Spec for Comet result - calculates the comet score for the given instance using the version specified in the spec. - `score` number, float: Output only. Comet score. Range depends on version. - `exactMatchResults` GoogleCloudAiplatformV1ExactMatchResults: Results for exact match metric. - `exactMatchMetricValues` GoogleCloudAiplatformV1ExactMatchMetricValue[]: Output only. Exact match metric values. - `score` number, float: Output only. Exact match score. - `fluencyResult` GoogleCloudAiplatformV1FluencyResult: Spec for fluency result. - `confidence` number, float: Output only. Confidence for fluency score. - `explanation` string: Output only. Explanation for fluency score. - `score` number, float: Output only. Fluency score. - `fulfillmentResult` GoogleCloudAiplatformV1FulfillmentResult: Spec for fulfillment result. - `confidence` number, float: Output only. Confidence for fulfillment score. - `explanation` string: Output only. Explanation for fulfillment score. - `score` number, float: Output only. Fulfillment score. - `groundednessResult` GoogleCloudAiplatformV1GroundednessResult: Spec for groundedness result. - `confidence` number, float: Output only. Confidence for groundedness score. - `explanation` string: Output only. Explanation for groundedness score. - `score` number, float: Output only. Groundedness score. - `metricResults` GoogleCloudAiplatformV1MetricResult[]: Metric results for each instance. The order of the metric results is guaranteed to be the same as the order of the instances in the request. - `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. - `explanation` string: Output only. The explanation for the metric result. - `rubricVerdicts` GoogleCloudAiplatformV1RubricVerdict[]: Output only. For rubric-based metrics, the verdicts for each rubric. - `evaluatedRubric` GoogleCloudAiplatformV1Rubric: Message representing a single testable criterion for evaluation. One input prompt could have multiple rubrics. - `content` GoogleCloudAiplatformV1RubricContent: Content of the rubric, defining the testable criteria. - `property` GoogleCloudAiplatformV1RubricContentProperty: Defines criteria based on a specific property. - `description` string: Description of the property being evaluated. Example: "The model's response is grammatically correct." - `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". - `reasoning` string: Optional. Human-readable reasoning or explanation for the verdict. This can include specific examples or details from the evaluated content that justify the given verdict. - `verdict` boolean: Required. Outcome of the evaluation against the rubric, represented as a boolean. `true` indicates a "Pass", `false` indicates a "Fail". - `score` number, float: Output only. The score for the metric. Please refer to each metric's documentation for the meaning of the score. - `metricxResult` GoogleCloudAiplatformV1MetricxResult: Spec for MetricX result - calculates the MetricX score for the given instance using the version specified in the spec. - `score` number, float: Output only. MetricX score. Range depends on version. - `pairwiseMetricResult` GoogleCloudAiplatformV1PairwiseMetricResult: Spec for pairwise metric result. - `customOutput` GoogleCloudAiplatformV1CustomOutput: Spec for custom output. - `rawOutputs` GoogleCloudAiplatformV1RawOutput: Raw output. - `rawOutput` string[]: Output only. Raw output string. - `explanation` string: Output only. Explanation for pairwise metric score. - `pairwiseChoice` 'PAIRWISE_CHOICE_UNSPECIFIED' | 'BASELINE' | 'CANDIDATE' | 'TIE': Output only. Pairwise metric choice. - `pairwiseQuestionAnsweringQualityResult` GoogleCloudAiplatformV1PairwiseQuestionAnsweringQualityResult: Spec for pairwise question answering quality result. - `confidence` number, float: Output only. Confidence for question answering quality score. - `explanation` string: Output only. Explanation for question answering quality score. - `pairwiseChoice` 'PAIRWISE_CHOICE_UNSPECIFIED' | 'BASELINE' | 'CANDIDATE' | 'TIE': Output only. Pairwise question answering prediction choice. - `pairwiseSummarizationQualityResult` GoogleCloudAiplatformV1PairwiseSummarizationQualityResult: Spec for pairwise summarization quality result. - `confidence` number, float: Output only. Confidence for summarization quality score. - `explanation` string: Output only. Explanation for summarization quality score. - `pairwiseChoice` 'PAIRWISE_CHOICE_UNSPECIFIED' | 'BASELINE' | 'CANDIDATE' | 'TIE': Output only. Pairwise summarization prediction choice. - `pointwiseMetricResult` GoogleCloudAiplatformV1PointwiseMetricResult: Spec for pointwise metric result. - `customOutput` GoogleCloudAiplatformV1CustomOutput: Spec for custom output. - `rawOutputs` GoogleCloudAiplatformV1RawOutput: Raw output. - `rawOutput` string[]: Output only. Raw output string. - `explanation` string: Output only. Explanation for pointwise metric score. - `score` number, float: Output only. Pointwise metric score. - `questionAnsweringCorrectnessResult` GoogleCloudAiplatformV1QuestionAnsweringCorrectnessResult: Spec for question answering correctness result. - `confidence` number, float: Output only. Confidence for question answering correctness score. - `explanation` string: Output only. Explanation for question answering correctness score. - `score` number, float: Output only. Question Answering Correctness score. - `questionAnsweringHelpfulnessResult` GoogleCloudAiplatformV1QuestionAnsweringHelpfulnessResult: Spec for question answering helpfulness result. - `confidence` number, float: Output only. Confidence for question answering helpfulness score. - `explanation` string: Output only. Explanation for question answering helpfulness score. - `score` number, float: Output only. Question Answering Helpfulness score. - `questionAnsweringQualityResult` GoogleCloudAiplatformV1QuestionAnsweringQualityResult: Spec for question answering quality result. - `confidence` number, float: Output only. Confidence for question answering quality score. - `explanation` string: Output only. Explanation for question answering quality score. - `score` number, float: Output only. Question Answering Quality score. - `questionAnsweringRelevanceResult` GoogleCloudAiplatformV1QuestionAnsweringRelevanceResult: Spec for question answering relevance result. - `confidence` number, float: Output only. Confidence for question answering relevance score. - `explanation` string: Output only. Explanation for question answering relevance score. - `score` number, float: Output only. Question Answering Relevance score. - `rougeResults` GoogleCloudAiplatformV1RougeResults: Results for rouge metric. - `rougeMetricValues` GoogleCloudAiplatformV1RougeMetricValue[]: Output only. Rouge metric values. - `score` number, float: Output only. Rouge score. - `rubricBasedInstructionFollowingResult` GoogleCloudAiplatformV1RubricBasedInstructionFollowingResult: Result for RubricBasedInstructionFollowing metric. - `rubricCritiqueResults` GoogleCloudAiplatformV1RubricCritiqueResult[]: Output only. List of per rubric critique results. - `rubric` string: Output only. Rubric to be evaluated. - `verdict` boolean: Output only. Verdict for the rubric - true if the rubric is met, false otherwise. - `score` number, float: Output only. Overall score for the instruction following. - `safetyResult` GoogleCloudAiplatformV1SafetyResult: Spec for safety result. - `confidence` number, float: Output only. Confidence for safety score. - `explanation` string: Output only. Explanation for safety score. - `score` number, float: Output only. Safety score. - `summarizationHelpfulnessResult` GoogleCloudAiplatformV1SummarizationHelpfulnessResult: Spec for summarization helpfulness result. - `confidence` number, float: Output only. Confidence for summarization helpfulness score. - `explanation` string: Output only. Explanation for summarization helpfulness score. - `score` number, float: Output only. Summarization Helpfulness score. - `summarizationQualityResult` GoogleCloudAiplatformV1SummarizationQualityResult: Spec for summarization quality result. - `confidence` number, float: Output only. Confidence for summarization quality score. - `explanation` string: Output only. Explanation for summarization quality score. - `score` number, float: Output only. Summarization Quality score. - `summarizationVerbosityResult` GoogleCloudAiplatformV1SummarizationVerbosityResult: Spec for summarization verbosity result. - `confidence` number, float: Output only. Confidence for summarization verbosity score. - `explanation` string: Output only. Explanation for summarization verbosity score. - `score` number, float: Output only. Summarization Verbosity score. - `toolCallValidResults` GoogleCloudAiplatformV1ToolCallValidResults: Results for tool call valid metric. - `toolCallValidMetricValues` GoogleCloudAiplatformV1ToolCallValidMetricValue[]: Output only. Tool call valid metric values. - `score` number, float: Output only. Tool call valid score. - `toolNameMatchResults` GoogleCloudAiplatformV1ToolNameMatchResults: Results for tool name match metric. - `toolNameMatchMetricValues` GoogleCloudAiplatformV1ToolNameMatchMetricValue[]: Output only. Tool name match metric values. - `score` number, float: Output only. Tool name match score. - `toolParameterKeyMatchResults` GoogleCloudAiplatformV1ToolParameterKeyMatchResults: Results for tool parameter key match metric. - `toolParameterKeyMatchMetricValues` GoogleCloudAiplatformV1ToolParameterKeyMatchMetricValue[]: Output only. Tool parameter key match metric values. - `score` number, float: Output only. Tool parameter key match score. - `toolParameterKvMatchResults` GoogleCloudAiplatformV1ToolParameterKVMatchResults: Results for tool parameter key value match metric. - `toolParameterKvMatchMetricValues` GoogleCloudAiplatformV1ToolParameterKVMatchMetricValue[]: Output only. Tool parameter key value match metric values. - `score` number, float: Output only. Tool parameter key value match score. - `trajectoryAnyOrderMatchResults` GoogleCloudAiplatformV1TrajectoryAnyOrderMatchResults: Results for TrajectoryAnyOrderMatch metric. - `trajectoryAnyOrderMatchMetricValues` GoogleCloudAiplatformV1TrajectoryAnyOrderMatchMetricValue[]: Output only. TrajectoryAnyOrderMatch metric values. - `score` number, float: Output only. TrajectoryAnyOrderMatch score. - `trajectoryExactMatchResults` GoogleCloudAiplatformV1TrajectoryExactMatchResults: Results for TrajectoryExactMatch metric. - `trajectoryExactMatchMetricValues` GoogleCloudAiplatformV1TrajectoryExactMatchMetricValue[]: Output only. TrajectoryExactMatch metric values. - `score` number, float: Output only. TrajectoryExactMatch score. - `trajectoryInOrderMatchResults` GoogleCloudAiplatformV1TrajectoryInOrderMatchResults: Results for TrajectoryInOrderMatch metric. - `trajectoryInOrderMatchMetricValues` GoogleCloudAiplatformV1TrajectoryInOrderMatchMetricValue[]: Output only. TrajectoryInOrderMatch metric values. - `score` number, float: Output only. TrajectoryInOrderMatch score. - `trajectoryPrecisionResults` GoogleCloudAiplatformV1TrajectoryPrecisionResults: Results for TrajectoryPrecision metric. - `trajectoryPrecisionMetricValues` GoogleCloudAiplatformV1TrajectoryPrecisionMetricValue[]: Output only. TrajectoryPrecision metric values. - `score` number, float: Output only. TrajectoryPrecision score. - `trajectoryRecallResults` GoogleCloudAiplatformV1TrajectoryRecallResults: Results for TrajectoryRecall metric. - `trajectoryRecallMetricValues` GoogleCloudAiplatformV1TrajectoryRecallMetricValue[]: Output only. TrajectoryRecall metric values. - `score` number, float: Output only. TrajectoryRecall score. - `trajectorySingleToolUseResults` GoogleCloudAiplatformV1TrajectorySingleToolUseResults: Results for TrajectorySingleToolUse metric. - `trajectorySingleToolUseMetricValues` GoogleCloudAiplatformV1TrajectorySingleToolUseMetricValue[]: Output only. TrajectorySingleToolUse metric values. - `score` number, float: Output only. TrajectorySingleToolUse score. ## Changes - **2026-11-06** `3aca3b348f0c`: 4 info - the security scope `https://www.googleapis.com/auth/aiplatform` was added to the endpoint's security scheme `Oauth2` - the security scope `https://www.googleapis.com/auth/aiplatform` was added to the endpoint's security scheme `Oauth2c` - the security scope `https://www.googleapis.com/auth/cloud-platform` was removed from the endpoint's security scheme `Oauth2` - the security scope `https://www.googleapis.com/auth/cloud-platform` was removed from the endpoint's security scheme `Oauth2c` - **2026-10-30** `34fcc627b062`: 4 info - the security scope `https://www.googleapis.com/auth/aiplatform` was added to the endpoint's security scheme `Oauth2` - the security scope `https://www.googleapis.com/auth/aiplatform` was added to the endpoint's security scheme `Oauth2c` - the security scope `https://www.googleapis.com/auth/cloud-platform` was removed from the endpoint's security scheme `Oauth2` - the security scope `https://www.googleapis.com/auth/cloud-platform` was removed from the endpoint's security scheme `Oauth2c` [Operation history](https://skmtc.dev/google/apis/aiplatform/changes/v1/projects/:projectsId/locations/:locationsId:evaluateInstances/post.md) --- [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)