Oracle Cloud Infrastructure Documentation

Overview of Generative AI Service

Oracle Cloud Infrastructure Generative AI is a fully managed service that provides a set of state-of-the-art, customizable large language models (LLMs) that cover a wide range of use cases for text generation.

Use the playground to try out the ready-to-use pretrained models, or create and host your own fine-tuned custom models based on your own data on dedicated AI clusters.

The OCI Generative AI service includes the following foundational models:

Generation
Give instructions to generate text or extract information from your text.
Summarization
Summarize text with your instructed format, length, and tone.
Embedding
Convert text to vector embeddings to use in applications for semantic searches, recommender systems, text classification, or text clustering.

Using the Pretrained Foundational Models

To get started, use the playground to try the pretrained foundational models. Run your prompts, adjust the parameters, update your prompts, and rerun the models until you're happy with the results. Then get the code from the Console and copy the code into your applications.

User workflow options for using the Generative AI service ready-to-use pretrained models.

Fine-tuning the Pretrained Models

You can create a copy of a pretrained foundational model, add your own training dataset, and let the OCI Generative AI service fine-tune the model for you. OCI Generative AI uses dedicated AI clusters specially sized for fine-tuning. These clusters belong only to your tenancy. After your model is fine-tuned, you create an endpoint for the custom model and host that model on a dedicated AI cluster that's designed for hosting. When you create the hosting cluster, select the correct pretrained model from which the fine-tuned model is derived from.

User workflow options for using the Generative AI service for fine-tuning a pretrained model and hosting the fine-tuned model through an endpoint.

Use Cases

Use the OCI Generative AI service for the following types of use cases.

Text Generation

Use the pretrained text generation models to create text for any purpose, for example:

  • Pitch for a new product
  • Slogan for a marketing campaign
  • Sales email to a client
  • Social media post
  • Job description
  • Title for an article
Conversation

You can ask questions in natural language about text submitted to the LLM—such as documents, emails, and product reviews—and the LLM reasons over the text and provides intelligent answers.

Data Extraction

Extract specific pieces of data from text, for example:

  • Extract applicant information from an application written in free-form text.
  • Extract dates or sums from a contract.
  • Extract insights or trends from data tables.
Summarization

Generate executive summaries for documents that are too long to read, or summarize any type of text, for example:

  • Documents
  • Contracts
  • Emails
  • Articles
  • Blog posts
  • Product reviews
  • Social media posts
Classification

Classify text into predefined categories, for example:

  • Given a list of support tickets, classify them by the department that should handle them.
  • Given a list of sectors and company names, classify the companies by their respective sectors.
Style Transfer

Change the style or tone of text, for example:

  • Rewrite any text in a different style, format (list or paragraphs), or tone.
  • Rephrase text.
  • Suggest grammatical improvements.
Semantic Similarity

Evaluate several inputs based on how similar their meaning is, for example:

  • Evaluate a list of questions sent to a support system to extract the most relevant answer given to similar questions in the past when a new question comes in.
  • Replace keyword-based searches with semantic searches to improve search results relevance.

Regions with Generative AI

Oracle hosts its OCI services in regions and availability domains. A region is a localized geographic area, and an availability domain is one or more data centers in that region. OCI Generative AI is hosted in the following region:

  • Region name: US Midwest (Chicago)
  • Region identifier: us-chicago-1

See About Regions and Availability Domains for the list of available OCI regions, along with associated locations, region identifiers, region keys, and availability domains.

Accessing Generative AI in the Console

  1. Sign in to the Console by using a supported browser.
  2. In the navigation bar of the Console, select a region with Generative AI, for example, US Midwest (Chicago). If you don't know which region to select, see Regions with Generative AI.
  3. Open the navigation menu and click Analytics & AI. Under AI Services, click Generative AI.