Memory
Redis-Backed Chat Memory
Memory
Redis-Backed Chat Memory
The Redis-Backed Chat Memory node is a component that provides long-term memory storage for chat conversations using Redis as the backend. It summarizes and stores conversation history, allowing for persistent and scalable chat memory across sessions.
Node Details
- Name: RedisBackedChatMemory
- Type: RedisBackedChatMemory
- Version: 2.0
- Category: Memory
Base Classes
- RedisBackedChatMemory
- BufferMemory (and its base classes)
Parameters
Credential
- Type: credential
- Credential Names: redisCacheApi, redisCacheUrlApi
- Description: Redis connection credentials
Credential
- Type: credential
- Credential Names: redisCacheApi, redisCacheUrlApi
- Description: Redis connection credentials
Inputs
-
Session Id (optional)
-
Session Id (optional)
- Type: string
- Default: Empty string
- Description: Unique identifier for the chat session. If not specified, a random id will be used
- Description: Unique identifier for the chat session. If not specified, a random id will be used
-
Session Timeouts (optional)
-
Session Timeouts (optional)
- Type: number
- Description: Time-to-live (TTL) for the session in seconds. Omit this parameter to make sessions never expire
- Description: Time-to-live (TTL) for the session in seconds. Omit this parameter to make sessions never expire
-
Memory Key
- Type: string
- Default: “chat_history”
- Description: Key used to store and retrieve the chat history in Redis
- Description: Key used to store and retrieve the chat history in Redis
-
Window Size (optional)
-
Window Size (optional)
- Type: number
- Description: Number of recent back-and-forth interactions to use as memory context
- Description: Number of recent back-and-forth interactions to use as memory context
Functionality
-
Initialization:
- Connects to Redis using provided credentials (either URL or individual connection parameters).
- Sets up a RedisChatMessageHistory instance for managing chat history.
- Creates a BufferMemoryExtended instance for handling memory operations.
-
Memory Operations:
- getChatMessages: Retrieves chat messages from Redis, with options for windowing and prepending messages.
- addChatMessages: Adds new messages (both user and AI) to the Redis store.
- clearChatMessages: Deletes all messages for a given session from Redis.
-
Session Management:
- Supports session-based storage using session IDs.
- Optional session timeout (TTL) for automatic expiration of old sessions.
Use Cases
- Long-running chatbots that need to maintain context across multiple interactions.
- Multi-user chat systems where each user’s history needs to be stored separately.
- Applications requiring scalable and persistent chat memory storage.
Integration
This node can be used in a AI solution to provide long-term memory capabilities to language models or other AI components that benefit from conversation history.
Notes
- Ensure that a Redis server is properly set up and accessible for this node to function correctly.
- When using in production, consider security implications and ensure proper authentication and encryption for Redis connections.
- The window size parameter can be used to limit the amount of history provided to AI models, which can be useful for managing token limits or focusing on recent context.