Tutorial

Getting started with vector sets

February 25, 202613 minute read

Prasan Rajpurohit

TL;DR:

Redis vector sets are a native data type that stores elements with associated vectors for fast similarity search. Use VADD to insert items with embeddings, and VSIM to find the nearest neighbors by comparing vectors. Vector sets power use cases like semantic search, recommendation systems, and AI-powered retrieval directly inside Redis.

#What you'll learn

What Redis vector sets are and when to use them
How to add elements with vectors using the VADD command
How to perform similarity search against existing elements or custom vectors using VSIM
How to bulk import vector data via Redis Insight
How to inspect vector sets with utility commands like VCARD, VDIM, and VINFO

#Prerequisites

Redis 8 or later with vector sets support enabled
Redis Insight (recommended for bulk import and exploring data)
Familiarity with generating vector embeddings (e.g., via OpenAI, Sentence Transformers, or similar models)

#What are Redis vector sets?

Vector sets are a Redis data type similar to sorted sets. However, instead of associating each element with a numerical score, elements in a vector set are associated with a vector—a list of floating-point numbers representing the item in a multi-dimensional space.

This makes vector sets ideal for similarity search tasks such as:

Retrieving the most similar items to the vector of an existing element already in the set.
Retrieving the most similar items to a specified vector (e.g., a new embedding not yet in the set).

With these capabilities, vector sets are useful in semantic search, recommendation systems, face recognition, and other apps where vector similarity is important.

A vector set is a collection of elements, each associated with a vector and optional custom attributes.

In the above representation:

Vector set key is "pg:sts"
Element IDs are sentences identified by "s1", "s2", etc., each storing a vector and custom attributes required for future filtering or inspection.

#How do you add items to a Redis vector set?

You can add items to a vector set using the VADD command.

#VADD syntax

Parameters:

{vectorSetKey} – The name of your vector set.
{embeddingDimension} – The length of the vector (number of dimensions).
{embeddingValues...} – The vector values (space-separated floats).
{elementId} – A unique identifier for this element in the set.
{elementAttributesAsJSON} – A JSON object containing any metadata about the element, making it easy to filter or inspect later.

#How do you bulk import vector data?

The full Semantic Textual Similarity (STS) Development Set used in this tutorial is available here.

Upload that file into Redis Insight:

Click **Bulk Actions** -> choose **Upload Data** tab -> upload the file and click **Upload**

Redis Insight Bulk Actions interface for uploading a data file

Post upload, you can see the status of the upload data

Redis Insight displaying the status and success of a bulk data upload

#How do you search by similarity with existing elements?

The VSIM command allows you to find elements in a vector set that are most similar to the vector of an existing element in the same set.

This is useful when you already have an element in your dataset and want to find others that are semantically or visually close to it—a common pattern in recommendation systems and "more like this" features.

#VSIM with existing elements

Parameters:

{vectorSetKey} – The name of your vector set.
ELE {elementId} – The ID of the element whose vector you want to use for similarity search.
WITHSCORES – Returns the similarity score for each result.
WITHATTRIBS – Returns the stored attributes (metadata) for each result.

#Try element similarity in the Redis sandbox

You can experiment with Element similarity queries in the Redis Sandbox:

Element similarity with scores and count example

Redis Sandbox environment showing results for an element similarity search with VSIM ELE command

Element similarity with logical filter

Redis Sandbox showing VSIM element similarity results filtered by activityType attribute

TIP

In the Redis sandbox, explore additional filter options in the left sidebar, including arithmetic filters, comparison filters, and containment filters.

#How do you search with a new vector (KNN query)?

The VSIM command can also search for elements similar to a vector you provide directly, instead of using an existing element's vector. This is a K-nearest neighbor (KNN) query.

This is useful when:

You have a new piece of text, image, or audio not in your dataset.
You have already generated its vector embeddings using the same model and dimensions used when seeding the vector set.

#VSIM with specified vectors

Where:

{vectorSetKey} – The name of your vector set.
VALUES {embeddingDimension} {embeddingValues...} – The embedding vector to compare against.
WITHSCORES – Returns the similarity score for each result.
WITHATTRIBS – Returns the stored attributes (metadata) for each result.
COUNT N (optional) – Limits the number of results returned.

NOTE

Before running this query, convert your search text into vector embeddings using the same model (and dimensionality) as the one used to seed the dataset. For example, if the dataset was built using OpenAI's text-embedding-ada-002 model (1536 dimensions), use the same for the query.

#Try value similarity in the Redis sandbox

Experiment with Value similarity queries in the Redis Sandbox:

Value similarity with scores and count example

Redis Sandbox showing KNN similarity search results for a provided query vector using VSIM VALUES

Value similarity with logical filter

Redis Sandbox showing VSIM value similarity results filtered with a logical attribute filter

TIP

In the Redis sandbox, explore additional filter options in the left sidebar, including arithmetic filters, comparison filters, and containment filters.

#What other vector set commands are available?

Vector sets in Redis support several utility commands that let you inspect, debug, and retrieve metadata, attributes, or structure-related information.

VCARD – Count elements.
VDIM – Get vector dimensions.
VEMB – Get vector for an element.
VGETATTR – Get attributes for an element.
VINFO – Get metadata about the vector set.
VISMEMBER – Check if an element exists.
VLINKS – Get neighbors in the HNSW graph.
VRANDMEMBER – Get random elements.

#Commands

#Try vector set commands in the Redis sandbox

Experiment with other vector set commands in the Redis sandbox:

Redis Sandbox interface showing VCARD, VDIM, and other vector set utility commands with output

TIP

In the Redis sandbox, you can select a command from the left sidebar, click the Run button, and instantly see the output.

#How do vector sets compare to vector indexing with Redis Query Engine?

Redis offers two ways to work with vectors:

Feature	Vector sets	Redis Query Engine vector indexing
Data type	Native `VSET` type (Redis 8+)	Vectors stored in hashes or JSON, indexed via `FT.CREATE`
Indexing	Built-in HNSW graph, automatic	Requires explicit index creation with `FT.CREATE`
Query command	`VSIM`	`FT.SEARCH` with KNN syntax
Filtering	Attribute-based filters in `VSIM`	Full-text, tag, numeric, and geo filters via query syntax
Best for	Lightweight similarity search, real-time recommendations, rapid prototyping	Complex queries combining vector search with structured filters
Schema	Schema-free; attributes are arbitrary JSON	Schema defined at index creation

Use vector sets when you need a simple, fast path to nearest-neighbor search with minimal setup. Use Redis Query Engine when you need to combine vector similarity with rich filtering, full-text search, or aggregation across structured data.

For a hands-on walkthrough using Redis Query Engine for vector search, see the Vector similarity search tutorial.

#Ready to use Redis vector sets?

You've learned how to:

Add elements to a vector set with VADD.
Run similarity searches using existing elements or custom query vectors with VSIM.
Perform KNN queries against new embeddings.
Use utility commands to inspect and explore your data.

Vector sets provide fast, scalable nearest-neighbor search natively in Redis — perfect for semantic search, recommendations, and AI-powered retrieval.

#Next steps

Try the Redis sandbox links throughout this guide.
Import your own data and test queries.
Explore the full vector sets data type documentation.
Learn about vector similarity search with Redis Query Engine.
Build a semantic text search application using Redis vectors.
Dive into the VADD and VSIM command references.