Valkey - Langflow — ContextQMD

import Icon from "@site/src/components/icon"; import PartialParams from '@site/docs/_partial-hidden-params.mdx'; import PartialConditionalParams from '@site/docs/_partial-conditional-params.mdx'; import PartialVectorSearchResults from '@site/docs/_partial-vector-search-results.mdx'; import PartialVectorStoreInstance from '@site/docs/_partial-vector-store-instance.mdx'; import { GraduatedBundleInstall } from '@site/docs/_partial-bundle-graduated-install.mdx';

<Icon name="Blocks" aria-hidden="true" /> Bundles contain custom components that support specific third-party integrations with Langflow.

This page describes the components that are available in the Valkey bundle.

Valkey is an open source, high-performance key/value datastore forked from Redis. It supports workloads such as caching, message queues, and can act as a primary database. Valkey is wire-compatible with Redis, so existing Redis clients and tools work with it out of the box.

Start the Valkey server

The Valkey Chat Memory component works with any Valkey server, with no additional modules required. To run a Valkey server with no vector store, run the standard valkey/valkey Docker image:

bash

docker run -d --name valkey -p 6379:6379 valkey/valkey:8

The Valkey Vector Store requires the valkey-search module for FT.CREATE and FT.SEARCH commands. The standard valkey/valkey Docker image does not include this module. Instead, use the valkey/valkey-bundle Docker image, which includes search enabled by default.

To start a Valkey server with search support, run:

bash

docker run -d --name valkey -p 6379:6379 valkey/valkey-bundle:8.1

Valkey Chat Memory component

The Valkey Chat Memory component retrieves and stores chat messages using a Valkey server.

Chat memories are passed between memory storage components as the Memory data type. The connection URL uses the redis:// scheme. Under the hood, this component uses the RedisChatMessageHistory class from langchain-community to maintain wire compatibility between Valkey and Redis.

For more information about using external chat memory in flows, see the Message History component.

Valkey Chat Memory parameters

Name	Display Name	Info
host	Hostname	Input parameter. The IP address or hostname of the Valkey server. Default: `localhost`.
port	Port	Input parameter. The Valkey port number. Default: `6379`.
database	Database	Input parameter. The Valkey database number. Default: `0`.
username	Username	Input parameter. The Valkey username (optional).
password	Valkey Password	Input parameter. The password for authentication (optional).
key_prefix	Key prefix	Input parameter. A prefix for message keys in Valkey (optional).
session_id	Session ID	Input parameter. The unique session identifier for the chat messages.

Valkey vector store

The Valkey vector store component reads and writes to Valkey vector stores using the ValkeyVectorStore class from langchain-aws.

This component uses the valkey-glide client library to communicate with the Valkey server and supports the FT.CREATE and FT.SEARCH commands for vector indexing and similarity search.

<details> <summary>About vector store instances</summary> <PartialVectorStoreInstance /> </details> <PartialVectorSearchResults />

:::tip For a tutorial using a vector database in a flow, see Create a vector RAG chatbot. :::

Valkey vector store parameters

You can inspect a vector store component's parameters to learn more about the inputs it accepts, the features it supports, and how to configure it.

For information about accepted values and functionality, see the Valkey documentation or inspect component code.

Name	Type	Description
valkey_server_url	SecretString	Input parameter. The Valkey server connection string (for example, `valkey://localhost:6379`).
valkey_index_name	String	Input parameter. The name of the Valkey vector index. Required when no documents are provided.
ingest_data	Data	Input parameter. The data to be ingested into the vector store.
search_query	String	Input parameter. The query for similarity search.
embedding	Embeddings	Input parameter. The embedding function to use.
number_of_results	Integer	Input parameter. The number of results to return in search. Default: `4`.
should_cache_vector_store	Boolean	Input parameter. If `true`, the component caches the vector store for the current build so multiple outputs can share it.