AI search

AI search streamlines your workflow by generating embeddings automatically. SmartObserve converts text to vectors during indexing and querying. It creates and indexes vector embeddings for documents and then processes query text into embeddings to find and return the most relevant results.

Prerequisite

Before using AI search, you must set up an ML model for embedding generation. When selecting a model, you have the following options:

Use a pretrained model provided by SmartObserve. For more information, see SmartObserve-provided pretrained models.
Upload your own model to SmartObserve. For more information, see Custom local models.
Connect to a foundation model hosted on an external platform. For more information, see Connecting to externally hosted models.

Tutorial

Getting started with semantic and hybrid search

Learn how to implement semantic and hybrid search

AI search methods

Once you set up an ML model, choose one of the following search methods.

Semantic search

Uses dense retrieval based on text embedding models to search text data.

Hybrid search

Combines keyword and semantic search to improve search relevance.

Multimodal search

Uses multimodal embedding models to search text and image data.

Neural sparse search

Uses sparse retrieval based on sparse embedding models to search text data.

Neural sparse ANN search

Uses approximate nearest neighbor techniques on sparse vectors for improved performance at scale.

Conversational search with RAG

Uses retrieval-augmented generation (RAG) and conversational memory to provide context-aware responses.

Prerequisite
Tutorial
AI search methods