HeatWave MySQL is the only fully managed MySQL database service that combines transactions, analytics, machine learning, and GenAI services, without ETL duplication. It also includes HeatWave Lakehouse, allowing users to query data stored in object storage,MySQL databases, or a combination of both. Users can deploy HeatWave MySQL–powered apps on a choice of public clouds: Oracle Cloud Infrastructure (OCI), Amazon Web Services (AWS), and Microsoft Azure.

This post shows you how to use free HeatWave GenAI features, in particular small LLMs that run in an Always Free HeatWave instance, for summarization, translation, chat, RAG, or other GenAI use cases. Additionally, you can use machine learning in conjunction with GenAI features and develop applications using JavaScript. All running on OCI Free Compute shapes—no GPUs needed and zero cost to get started.

All OCI accounts have access to a HeatWave Always Free instance in their OCI home region. This lets organizations experience one or more  of the latest HeatWave features free-of charge. We have added small LLMs to the Always Free instance so that customers can expand their applications to use GenAI.

Customers can develop and run applications accessing data from Object Storage and MySQL database using these features. Additionally, they can also build machine learning models with HeatWave AutoML and see how it saves considerable time for developers and DBAs.

 

Use cases for In-HeatWave LLM Capabilities

These new in-HeatWave LLMs enable customers to use HeatWave GenAI along with HeatWave Vector store to build new applications and then deploy them.

Generate New Content

Enables enterprises to create content in multiple languages for activities such as  social media posts, blog articles, and email campaigns.

Summarize Content

Enterprises can effortlessly generate concise and correct summaries of documents, reports, and logs while keeping essential information. For example, ecommerce sites with many SKUs—hundreds or thousands or more—have multiple reviews per product. Pouring over hundreds of reviews to get the gist of them is a laborious process. Users can tap into HeatWave GenAI’s capabilities to quickly summarize, translate, or even analyze sentiment in the reviews. Watch a demo or try it yourself.

Retrieval augmented Generation (RAG)

Enterprises can now use HeatWave GenAI to easily search their own documents or generate new content using the LLMs with HeatWave Vector Store to create contextual, business-specific applications. They can store and search proprietary documents using vector similarity search and augment LLM prompts with relevant business content—easily and securely. These techniques enable enterprises to build chatbots for a variety of needs such as answering questions about company policies, helping customers choose proper training courses etc. Watch a demo or try it yourself.
 

HeatWave GenAI in Always Free


Advantages of HeatWave GenAI

Use of HeatWave GenAI enables the applications to use state of the art LLMs at no extra cost, minimizing the need for human intervention for routine, repetitive activities. Note that since the LLMs are hosted within the database and accessible using simple to use APIs, there is no need to transfer your proprietary documents to a third party application or subscribe to external APIs, provision GPUs, or manage third-party integrations – thus improving the security for the data and application and minimizing the lead time for application development.

The new in-HeatWave LLMs bring generative AI to every developer—without added costs or complexity. One can prototype powerful AI applications, enhance business workflows, and deliver richer user experiences directly from your Free Tier environment.

  • No separate subscriptions
  • No data movement or egress
  • No ML expertise needed

It’s simple, fast, and completely free to get started.

Note that these models are also available in HeatWave on AWS. For more information, see the HeatWave on AWS GenAI documentation and HeatWave Vector Store on AWS.
 

Get Started

If you already have an OCI account, simply create a HeatWave MySQL DB System with the “Always Free” option and start exploring the capabilities. LLMs are preconfigured and ready to use within your free HeatWave instance. HeatWave GenAI now supports Llama 3.2-1B-Instruct and Llama 3.2-3B -Instruct. Both models are designed for execution on CPUs and low memory consumption.

New to OCI? Sign up for the Free Tier to get started.

For more details and technical information, see:

Start building with GenAI today using the OCI Free Tier. Discover the simplicity, speed, and power of HeatWave GenAI—on Always Free.