AI LangChain4j - List vs ChatMemory in LangChain4j

We can use a simple List<ChatMessage> to create a multi-turn conversation (check out this example), but that is not same as using ChatMemory. In this tutorial we are going to see what the difference is.

Core Difference

List<ChatMessage> is just a data container, while ChatMemory is a managed conversation state system.

List<ChatMessage> (Manual Control)

A simple collection you manage completely yourself.

Your Responsibilities:

• Adding messages in correct order
• Removing old messages manually
• Preventing unlimited growth
• Managing session separation

Limitations:

• No automatic message trimming
• Easy to exceed token limits
• No built-in strategies
• No persistence support

ChatMemory (Managed System)

A contract defining how conversation history is managed.

Advantages:

• Automatic message windowing
• Pluggable strategies (window, token-based, persistent)
• Built-in safety controls
• Session-aware design

When to Use Each

Use List<ChatMessage> for:

• Single-turn or very short conversations
• Demos and examples
• When you need absolute manual control

Use ChatMemory for:

• Real applications
• Multi-turn conversations
• Multiple users/sessions
• When you need persistence or scaling

Bottom Line

List<ChatMessage> is simple storage. ChatMemory is storage plus policy, lifecycle, and safety management.