HeatWave Release Notes
HeatWave GenAI is a feature of HeatWave that lets you communicate with unstructured data in HeatWave using natural-language queries. It uses a familiar SQL interface which makes it is easy to use for content generation, summarization, and retrieval-augmented generation (RAG).
Using HeatWave GenAI, you can perform natural-language searches in a single step using either in-database or external large language models (LLMs). All the elements that are necessary to use HeatWave GenAI with proprietary data are integrated and optimized to work with each other.
Note
This chapter assumes that you are familiar with HeatWave Database Systems.
HeatWave GenAI includes the following:
-
In-Database and OCI Generative AI Service LLMs
HeatWave GenAI uses large language models (LLMs) to enable natural language communication in multiple languages. You can use the capabilities of the LLMs to search data as well as generate or summarize content. However, as these LLMs are trained on public data, the responses to your queries are generated based on information available in the public data sources. To produce more relevant results, you can use the LLM capabilities of HeatWave GenAI with the vector store functionality to perform a vector search using RAG.
-
In-Database Vector Store
HeatWave GenAI provides an inbuilt vector store that you can use to store enterprise-specific proprietary content and perform vector or similarity across documents. Queries that you ask are automatically encoded with the same embedding model as the vector store without requiring any additional inputs or running a separate service. The vector store also provides valuable context for LLMs for RAG use cases.
-
Retrieval-Augmented Generation
HeatWave GenAI retrieves content from the vector store and provides it as context to the LLM along with the query. This process of generating an augmented prompt is called RAG, and it helps HeatWave GenAI produce more contextually relevant, personalised, and accurate results.
-
HeatWave Chat
This is an inbuilt chatbot that extends the LLMs capabilities as well as vector store and RAG functionalities of HeatWave GenAI to let you ask multiple follow-up questions about a topic in a single session. You can use HeatWave Chat to build customized chat applications by specifying custom settings, prompt, chat history length, and number of citations to be used for generating a response.
HeatWave Chat also provides a graphical interface integrated with the Visual Studio Code plugin for MySQL Shell.
HeatWave GenAI lets you integrate generative AI into the applications, providing an integrated end-to-end pipeline including vector store generation, vector search using RAG, and an inbuilt chatbot.
Some key benefits of using HeatWave GenAI are described below:
The natural-language processing (NLP) capabilities of the LLMs let non-technical users have human-like conversations with the system in natural language.
The in-database integration of LLMs and embedding generation eliminates the need for using external solutions, and ensures the security of the proprietary content.
The in-database integration of LLMs, vector store, and embedding generation simplifies complexity of applications that use these features.
The cost of running natural-language queries is significantly low as HeatWave GenAI is available at no additional cost for HeatWave users.
HeatWave GenAI integrates with other in-database capabilities such as machine learning, analytics, and Lakehouse.