Generative AI Leader Google Cloud's Generative AI Offerings — All Questions With Answers

Refer to the exhibit.

```json
{
  "instances": [
    {"content": "Translate to French: Hello, how are you?"}
  ],
  "parameters": {
    "temperature": 0.7,
    "maxOutputTokens": 100,
    "topP": 0.9
  }
}
```

A data scientist sends this request to a Gemini model endpoint and receives a response in English. What is the most likely reason?

Refer to the exhibit.

```
{
  "predictions": [
    {
      "content": "The capital of France is Paris.",
      "safetyAttributes": {
        "categories": ["Toxic"],
        "scores": [0.02],
        "blocked": false
      }
    }
  ],
  "metadata": {
    "tokenCount": 8
  }
}
```

Refer to the exhibit.

```
# deployment.yaml
apiVersion: serving.kserve.io/v1beta1
kind: InferenceService
metadata:
  name: my-model
spec:
  predictor:
    containers:
      - name: model-container
        image: us-central1-docker.pkg.dev/my-project/my-repo/my-model:latest
        resources:
          limits:
            nvidia.com/gpu: 1
```

{
  "bindings": [
    {
      "role": "roles/aiplatform.admin",
      "members": ["user:alice@example.com"]
    }
  ]
}

agent:
  display_name: travel_agent
  dynamic_actions:
    - action_name: book_flight
      http_endpoint: https://api.example.com/flights

Refer to the exhibit. {
  "name": "projects/123/locations/us-central1/batchPredictionJobs/bpj456",
  "model": "projects/123/locations/us-central1/models/789",
  "inputConfig": {
    "instancesFormat": "jsonl",
    "gcsSource": {"uris": ["gs://bucket/input.jsonl"]}
  },
  "outputConfig": {
    "predictionsFormat": "jsonl",
    "gcsDestination": {"outputUriPrefix": "gs://bucket/output/"}
  },
  "machineType": "n1-standard-4",
  "batchSize": 64,
  "startingReplicaCount": 1,
  "maxReplicaCount": 1
}

Refer to the exhibit.

Error: 400 INVALID_ARGUMENT: The request was invalid because the grounding configuration specifies a data store 'projects/123/locations/global/dataStores/my-ds' that does not exist.

Refer to the exhibit.
```gcloud output
$ gcloud ai models describe --region=us-central1 my-model@123
Model:
  displayName: my-model
  versionId: "123"
  modelSource: MODEL_GARDEN
  supportedExportFormats:
  - id: "json"
  supportedInputStorageFormats:
  - json
  supportedOutputStorageFormats:
  - json
```

Refer to the exhibit.
```json
{
  "predictions": [
    {
      "safetyAttributes": [
        {
          "categories": ["Toxicity", "Insult"],
          "scores": [0.85, 0.72]
        }
      ]
    }
  ],
  "deployedModelId": "123",
  "model": "projects/my-project/locations/us-central1/models/my-model",
  "modelDisplayName": "my-model"
}
```

Refer to the exhibit.
```
Deploying a model to Vertex AI Endpoint with GPU:
$ gcloud ai endpoints deploy-model $ENDPOINT_ID \
  --model=$MODEL_ID \
  --machine-type=n1-standard-4 \
  --accelerator=count=1,type=nvidia-tesla-t4 \
  --min-replica-count=2 \
  --max-replica-count=5
```

Refer to the exhibit.

{
  "name": "projects/123456789/locations/us-central1/endpoints/my-endpoint",
  "deployedModels": [
    {
      "id": "model-v1",
      "model": "projects/123456789/locations/us-central1/models/sentiment-model",
      "displayName": "Version 1",
      "trafficSplit": {"model-v1": 100}
    },
    {
      "id": "model-v2",
      "model": "projects/123456789/locations/us-central1/models/sentiment-model-v2",
      "displayName": "Version 2",
      "trafficSplit": {"model-v2": 0}
    }
  ],
  "trafficSplit": {"model-v1": 100, "model-v2": 0}
}

This is the output of `gcloud ai endpoints describe my-endpoint --region=us-central1`. You want to gradually shift 20% of traffic from model-v1 to model-v2.