Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

Currently, when enabling the memory function in LLM nodes, the system records the user's conversation history and uses it as a reference for the LLM’s prompts. However, in a single conversation flow, multiple LLM nodes may be involved, yet the memory is shared across all LLM nodes, rather than being stored separately for each node.

This design leads to the following issues:

Memory Sharing Issue: Different LLM nodes handle different tasks, but they share the same memory, which may cause incorrect references in responses.
Incorrect Responses: When switching between different LLM nodes, the memory from the previous LLM node may affect the next node’s responses, leading to unintended or incorrect answers.

2. Additional context or comments

We propose an enhancement to the memory mechanism of LLM nodes, ensuring that different LLM nodes maintain independent memories or allowing the option to clear memory when needed. The key requests are as follows:

1️⃣ Independent Memory Storage (Preferred)
Each LLM node should have its own isolated memory space to ensure no cross-contamination between nodes.
A conversation history from Node A should not influence responses from Node B.
2️⃣ Memory Reset Functionality
Allow manual or configurable options to clear an LLM node’s memory within a workflow.
Provide an API or UI option to reset memory for specific LLM nodes to prevent error propagation.
3️⃣ Configurable Memory Management
Allow users to choose the memory mode in LLM node settings, such as:
✅ Independent Memory Mode (Each LLM node retains separate memory)
✅ Shared Memory Mode (Current design, suitable for specific use cases)
✅ Auto-Clear Memory (Automatically clears memory after node execution)

3. Can you help us with this feature?

• I am interested in contributing to this feature.