Vertex AI → Agent Platform
PMLE name-mapping cheat sheet · Checked
Google's June 1, 2026 exam guide uses Gemini Enterprise Agent Platform names. Use this sheet to read older Vertex AI material alongside it. A rename does not establish identical capabilities, prices, endpoints or resource IDs.
Verified name mappings
| Old name | New name | What it’s for | Official doc |
|---|---|---|---|
| Vertex AI Platform | Gemini Enterprise Agent Platform | Build, train and manage ML models. | Google table |
| Vertex AI API | Gemini Enterprise Agent Platform API | API surface; this is a naming reference, not an endpoint migration guide. | Google table |
| Vertex AI Model Garden | Gemini Enterprise Agent Platform Model Garden | Find and select a model for a task. | Google table |
| Vertex AI Training | Gemini Enterprise Agent Platform Managed Training | Run managed model training. | Google table |
| Vertex AI Serverless Training | Gemini Enterprise Agent Platform Serverless Training | Use managed compute for distributed training. | Google table |
| Vertex AI Pipelines | Gemini Enterprise Agent Platform Pipelines | Build and orchestrate ML workflows. | Google table |
| Vertex AI Workbench | Gemini Enterprise Agent Platform Workbench | Prototype models in notebooks. | Google table |
| Vertex AI Feature Store | Feature Store on Gemini Enterprise Agent Platform | Manage and serve model features. | Google table |
| Vertex AI Model Registry | Model Registry on Gemini Enterprise Agent Platform | Organize and version models. | Google table |
| Vertex AI Model Evaluation | Model Evaluation on Gemini Enterprise Agent Platform | Evaluate model quality. | Google table |
| Vertex AI Experiments | Experiments on Agent Platform | Track and compare experiments, artifacts and versions. | Google table |
| Vertex AI Model Monitoring | Model Monitoring on Gemini Enterprise Agent Platform | Monitor production models and data drift. | Google table |
| Vertex AI AutoML (Vision/Video/Tables) | AutoML on Gemini Enterprise Agent Platform | Develop models with low-code tools; check current support before use. | Google table |
| Vertex AI Batch Inference | Gemini Enterprise Agent Platform Batch Inference | Run batch model inference. | Google table |
| Vertex AI Online Inference | Gemini Enterprise Agent Platform Online Inference | Serve online model inference. | Google table |
- PreviouslyVertex AI PlatformCurrent nameGemini Enterprise Agent Platform
Build, train and manage ML models.
Verify in Google’s naming table →- PreviouslyVertex AI APICurrent nameGemini Enterprise Agent Platform API
API surface; this is a naming reference, not an endpoint migration guide.
Verify in Google’s naming table →- PreviouslyVertex AI Model GardenCurrent nameGemini Enterprise Agent Platform Model Garden
Find and select a model for a task.
Verify in Google’s naming table →- PreviouslyVertex AI TrainingCurrent nameGemini Enterprise Agent Platform Managed Training
Run managed model training.
Verify in Google’s naming table →- PreviouslyVertex AI Serverless TrainingCurrent nameGemini Enterprise Agent Platform Serverless Training
Use managed compute for distributed training.
Verify in Google’s naming table →- PreviouslyVertex AI PipelinesCurrent nameGemini Enterprise Agent Platform Pipelines
Build and orchestrate ML workflows.
Verify in Google’s naming table →- PreviouslyVertex AI WorkbenchCurrent nameGemini Enterprise Agent Platform Workbench
Prototype models in notebooks.
Verify in Google’s naming table →- PreviouslyVertex AI Feature StoreCurrent nameFeature Store on Gemini Enterprise Agent Platform
Manage and serve model features.
Verify in Google’s naming table →- PreviouslyVertex AI Model RegistryCurrent nameModel Registry on Gemini Enterprise Agent Platform
Organize and version models.
Verify in Google’s naming table →- PreviouslyVertex AI Model EvaluationCurrent nameModel Evaluation on Gemini Enterprise Agent Platform
Evaluate model quality.
Verify in Google’s naming table →- PreviouslyVertex AI ExperimentsCurrent nameExperiments on Agent Platform
Track and compare experiments, artifacts and versions.
Verify in Google’s naming table →- PreviouslyVertex AI Model MonitoringCurrent nameModel Monitoring on Gemini Enterprise Agent Platform
Monitor production models and data drift.
Verify in Google’s naming table →- PreviouslyVertex AI AutoML (Vision/Video/Tables)Current nameAutoML on Gemini Enterprise Agent Platform
Develop models with low-code tools; check current support before use.
Verify in Google’s naming table →- PreviouslyVertex AI Batch InferenceCurrent nameGemini Enterprise Agent Platform Batch Inference
Run batch model inference.
Verify in Google’s naming table →- PreviouslyVertex AI Online InferenceCurrent nameGemini Enterprise Agent Platform Online Inference
Serve online model inference.
Verify in Google’s naming table →
Omitted: Agent Engine / Agent Runtime and the two Vector Search / Agent Retrieval rows. Their names are verified, but their capability descriptions were not independently rechecked for this sheet. No unverified mapping is included.
Six June 2026 PMLE domains
- 1. Architecting low-code AI solutions~13%
- 2. Collaborating within and across teams to manage data and models~16%
- 3. Scaling prototypes into ML models~21%
- 4. Serving and scaling models~20%
- 5. Automating and orchestrating ML pipelines~18%
- 6. Monitoring AI solutions~13%
Google's approximate weights total 101%. They are preserved here, not normalized or converted into question counts. June 1 is the guide's “as of” date, not a claimed exam cutover date.
Official sources
Google's Vertex AI name-change table · ML platform overview · June 1, 2026 PMLE exam guide. Retrieved 2026-10-17.
Print reference: docs.cloud.google.com/gemini-enterprise-agent-platform/vertex-ai-name-changes
This sheet is a study reference, not an exam-outcome guarantee.