Loss of multi round dialogue function

[qhkabuqiluo]

Self Checks

• This is only for bug report, if you would like to ask a quesion, please head to Discussions.
• 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).
• Pleas do not modify this template :) and fill in all the required fields.

Dify version

0.6.7

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

1. Create chatflow
2. Enable memory function
3. What is the communication subsidy for employee X for the first time
4. Second question for employee Y
   
   
   But I can still do it on version 0.6.4

✔️ Expected Behavior

Capable of conducting multiple rounds of dialogue

❌ Actual Behavior

No response

[dosubot]

To address the issue with the multi-round dialogue function not working as expected in Dify version 0.6.7, let's focus on potential solutions and checks:

1. Check Memory Settings: Verify that the memory settings are correctly configured. Ensure memory is enabled and the window size is set appropriately for your use case.

2. Model Providers Compatibility: Confirm that the model provider you're using supports Chinese language inputs effectively. Dify supports several providers optimized for Chinese, so ensure your choice is among them.

3. Review Logs for Insights: Examine Dify's logs for any error messages or unusual behavior that could indicate what's causing the issue with the multi-round dialogue function.

These steps are derived from understanding how Dify's LLMNode class and memory function configurations work, as well as the importance of choosing the right model provider for language support. If after these checks the issue still exists, providing more detailed information such as error logs or specific behaviors you're observing could help in diagnosing the problem more accurately.

Details

Sources

• en/features/workflow/node/llm.md
• web/app/components/workflow/nodes/_base/components/memory-config.tsx
• en/features/logs.md
• zh_CN/guides/logs.md
• .github/ISSUE_TEMPLATE/bug_report.yml
• zh_CN/getting-started/readme/model-providers.md
• sdks/nodejs-client/index.js
• api/core/workflow/nodes/llm/llm_node.py

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.