Skip to main content

Using Machine Learning Models in Weaviate

Overview​

Weaviate leverages two fundamental types of machine learning models to power AI-native applications:

  1. Embedding Models - Transform data into high-dimensional vector representations
  2. Generative Models - Create new content based on input prompts and context

This guide will help you understand how these models can be set up in Weaviate, briefly covering the functioning of these models on a high level as well.

We will look at how to use the search enabled by the two types of embedding models supported in Weaviate; Text embeddings and Multimodal embeddings.

This guide will also explore practical applications ranging from semantic search to agentic RAG applications.

Prerequisites​

  • A Node.js environment with weaviate-client installed
  • Basic understanding of Weaviate's search functionality
  • Intermediate JavaScript programming skills
  • You must have completed the quickstart

Learning objectives​

  What are these?
  Each unit includes a set of Learning Goals and Learning Outcomes which form the unit's guiding principles.
  • Learning Goals describe the unit's key topics and ideas.
  • Learning Outcomes on the other hand describe tangible skills that the learner should be able to demonstrate

  Here, we will cover:

Learning Goals
  • A high level understanding of embedding and generative models.
  • Distinguish between text and multimodal embedding types.
  • Configuring Weaviate to use embedding and generative models.
  • Making semantic and generative searches in Weaviate using JavaScript.

  By the time you are finished, you will be able to:

Learning Outcomes
  • Differentiate between embedding and generative machine learning models.
  • Configure Weaviate to use text and multimodal embedding models for semantic search.
  • Configure Weaviate to use supported generative models for generative search.

Questions and feedback​

If you have any questions or feedback, let us know in the user forum.