Generative AI
OCI Generative AI is a managed service for building generative AI applications with hosted models, custom models, endpoints, and dedicated AI clusters. This subject area enables tracking Generative AI resources across tenancies, compartments, regions, and time, with detail for dedicated AI clusters, endpoints, private endpoints, and models.
Business Questions:
The Subject Area can answer following business questions:
- How many Generative AI models, endpoints, and dedicated AI clusters exist by tenancy, compartment, or region?
- How are model, endpoint, and dedicated AI cluster counts changing over time?
- Which dedicated AI clusters are deployed by type and unit shape?
- Which endpoints have content moderation enabled?
- Which models are available by model type, vendor, version, or long-term support status?
Logical Model
The model centers on three visible fact folders for dedicated AI clusters, endpoints, and models.
Generative AI Dedicated AI Cluster
Generative AI Endpoint
Generative AI Model
Metric Details
| Fact Folder | Metric | Definition |
|---|---|---|
| Fact - Generative AI Dedicated AI Cluster | Unit Count | SUM visible presentation metric. |
| Fact - Generative AI Dedicated AI Cluster | Generative AI Dedicated AI Cluster Count | COUNT visible presentation metric. |
| Fact - Generative AI Endpoint | Generative AI Endpoint Count | COUNT visible presentation metric. |
| Fact - Generative AI Model | Resource Count | SUM visible presentation metric. |
| Fact - Generative AI Model | Generative AI Model Count | COUNT visible presentation metric. |