Enterprise AI Models in OCI Generative AI

Use Enterprise AI Models in OCI Generative AI to access pretrained hosted models, import supported custom models, and deploy models for enterprise inference workloads.

This section provides links to the models available in OCI Generative AI and to the resources used to deploy, manage, and scale those models in OCI.

What You Can Do with Enterprise AI Models

Use Enterprise AI Models when you want to:

  • Run inference with pretrained hosted models
  • Import and host supported custom models
  • Select on-demand and dedicated deployment options
  • Deploy models on dedicated AI clusters for production workloads
  • Manage endpoints and private network access
  • Review model and regional availability
  • Understand pricing and performance considerations

OCI Generative AI supports core model tasks such as:

  • Chat for conversational generation
  • Embeddings for semantic search, recommendation, classification, and clustering
  • Rerank for ordering documents by relevance to a query

Model Usage Options

OCI Generative AI supports multiple ways to use models:

  • Pretrained hosted models for managed inference through OCI
  • Imported models for supported custom model deployment
  • On-demand mode for shared managed access
  • Dedicated mode for isolated model serving on dedicated AI clusters

These options let you move from experimentation to production while selecting the level of control, performance isolation, and infrastructure management that fits your workload.

Model Infrastructure and Management

Enterprise AI Models in OCI Generative AI are supported by deployment and management resources such as:

  • Dedicated AI Clusters for isolated model hosting
  • Endpoints for serving model traffic
  • Private Endpoints for secure network access
  • Regional model availability for deployment planning
  • Performance and cost guidance for production workloads

Topics in This Section

Use the following topics to learn about Enterprise AI Models in OCI Generative AI: