Oracle Cloud Infrastructure Documentation

Overview of the Generative AI Service

OCI Generative AI is a fully managed Oracle Cloud Infrastructure service for building, deploying, and operating generative AI applications at enterprise scale.

Core Features

OCI Generative AI supports these core generative AI use cases:

  • Chat for conversational experiences such as question answering and virtual assistants
  • Embeddings for semantic search, recommendation, classification, and clustering
  • Rerank for ordering documents by relevance to a query
  • OpenAI-compatible APIs for integrating with existing tools and SDKs

Model Usage Options

You can use OCI Generative AI in these ways:

  • Use pretrained hosted models through the playground, API, or CLI.
  • Import, fine-tune, and host custom models on dedicated AI clusters.
  • Move from experimentation to production with enterprise controls.

Enterprise AI features

OCI Generative AI includes Enterprise AI features for building production-grade agentic applications.

These features include:

  • OCI Responses API
    • OpenAI Responses-compatible API for model interaction and agentic workflows
    • Supports orchestration, reasoning, tool use, memory, and multi-model routing
  • Tools
    • File Search
    • Code Interpreter
    • Function Calling for local tools
    • MCP Calling for remote MCP servers
    • Containers API
    • Vector Stores API
    • Files API
  • Memory
    • Conversations API
    • Long-term memory
    • Short-term memory context compaction
  • Projects
    • Organize agent workloads by project
    • Isolate conversations, files, containers, and memory
    • Configure data retention and memory settings
  • Applications
    • Fully managed hosting for agentic applications
    • Support for applications built with open source frameworks or MCP servers
    • Built-in security controls
    • Public and private endpoint support
  • Vector Stores
    • Managed vector storage
    • File ingestion
    • Semantic search
    • Metadata filtering
    • Support for RAG and NL2SQL use cases
  • NL2SQL
    • Ingests customer schema information
    • Enriches schema data into a semantic vector store
    • Accepts natural language queries and produces SQL
    • Runs in a permission-controlled manner without moving or copying database content
  • Enterprise AI API Keys
    • OCI-specific API keys for Enterprise AI services
    • Automatic rotation

Platform Benefits

  • Build production-ready AI applications faster
  • Reduce operational complexity
  • Apply enterprise governance and security controls
  • Use a unified platform for generative AI, retrieval, memory, tools, and managed hosting