io.lettuce.core

Class VAddArgs

java.lang.Object

io.lettuce.core.VAddArgs

All Implemented Interfaces:

CompositeArgument

public class VAddArgs
extends Object
implements CompositeArgument

Argument list builder for the Redis VADD command.

This class provides options for configuring how vectors are stored and indexed, including quantization type, exploration factor, attributes, and graph connectivity parameters.

Example usage:

 
 VAddArgs args = VAddArgs.checkAndSet(true)
                     .quantizationType(QuantizationType.Q8)
                     .explorationFactor(200)
                     .attributes("{\"name\": \"Point A\", \"description\": \"First point added\"}"));
 
 

VAddArgs is a mutable object and instances should be used only once to avoid shared mutable state.

Since:

6.7

Author:

Tihomir Mateev

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	VAddArgs.Builder Builder entry points for VAddArgs.

Constructor Summary

Constructors

Constructor and Description
VAddArgs()

Method Summary

Modifier and Type	Method and Description
VAddArgs	attributes(JsonValue attributes) Set the attributes for the vector (SETATTR).
VAddArgs	attributes(String attributes) Set the attributes for the vector (SETATTR).
<K,V> void	build(CommandArgs<K,V> args) Builds the command arguments based on the options set in this VAddArgs instance.
VAddArgs	checkAndSet(boolean checkAndSet) Set the CAS (Check-And-Set) flag for the vector addition.
VAddArgs	explorationFactor(Long explorationFactor) Set the exploration factor (EF) for the vector search.
VAddArgs	maxNodes(Long maxNodes) Set the maximum number of connections per node (M).
VAddArgs	quantizationType(QuantizationType quantType) Set the quantization type for the vector.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

VAddArgs

public VAddArgs()

Method Detail

checkAndSet

public VAddArgs checkAndSet(boolean checkAndSet)

Set the CAS (Check-And-Set) flag for the vector addition.

The CAS option performs the operation partially using threads, in a check-and-set style. The neighbor candidates collection, which is slow, is performed in the background, while the command is executed in the main thread.

This can improve performance when adding vectors to large sets.

Parameters:

checkAndSet - whether to enable the check-and-set operation mode.

Returns:

this

quantizationType

public VAddArgs quantizationType(QuantizationType quantType)

Set the quantization type for the vector.

Quantization affects how vectors are stored and impacts memory usage, performance, and recall quality:

• Q8 (default) - Uses signed 8-bit quantization, balancing memory usage and recall quality
• NOQUANT - Stores vectors without quantization, using more memory but preserving full precision
• BIN - Uses binary quantization, which is faster and uses less memory, but impacts recall quality

Note that these options are mutually exclusive.

Parameters:

quantType - the quantization type for the vector.

Returns:

this

explorationFactor

public VAddArgs explorationFactor(Long explorationFactor)

Set the exploration factor (EF) for the vector search.

The EF option plays a role in the effort made to find good candidates when connecting the new node to the existing Hierarchical Navigable Small World (HNSW) graph.

The default is 200. Using a larger value may help in achieving a better recall, but will increase the time needed to add vectors to the set.

Parameters:

explorationFactor - the exploration factor for the vector search (default: 200).

Returns:

this

maxNodes

public VAddArgs maxNodes(Long maxNodes)

Set the maximum number of connections per node (M).

The M option specifies the maximum number of connections that each node of the graph will have with other nodes.

The default is 16. More connections means more memory, but provides for more efficient graph exploration. Nodes at layer zero (every node exists at least at layer zero) have M * 2 connections, while the other layers only have M connections.

If you don't have a recall quality problem, the default is acceptable and uses a minimal amount of memory.

Parameters:

maxNodes - the maximum number of connections per node (default: 16).

Returns:

this

attributes

public VAddArgs attributes(String attributes)

Set the attributes for the vector (SETATTR).

The SETATTR option associates attributes in the form of a JavaScript object to the newly created entry or updates the attributes (if they already exist). It is the same as calling the VSETATTR command separately.

Attributes can be used for filtering during similarity searches with the VSIM command.

Example: attributes(Arrays.asList("{\"color\":\"red\",\"price\":25}"))

Parameters:

attributes - the attributes for the vector as JSON strings.

Returns:

this

attributes

public VAddArgs attributes(JsonValue attributes)

Set the attributes for the vector (SETATTR). Works with a JsonValue object.

The SETATTR option associates attributes in the form of a JavaScript object to the newly created entry or updates the attributes (if they already exist). It is the same as calling the VSETATTR command separately.

Attributes can be used for filtering during similarity searches with the VSIM command.

Parameters:

attributes - the attributes for the vector as JSON strings.

Returns:

this

build

public <K,V> void build(CommandArgs<K,V> args)

Builds the command arguments based on the options set in this VAddArgs instance.

This method is called internally by the Redis client to construct the VADD command with all the specified options. It adds each option that has been set to the command arguments in the correct format.

Specified by:

build in interface CompositeArgument

Type Parameters:

K - Key type.

V - Value type.

Parameters:

args - Command arguments to which the VADD options will be added.