langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-22 20:45:26 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

Loss of Chat History Retention in GLM Text Generation Model When Prompts Exceed 1000 Words (Agent) #6693

New Issue
Closed
opened 2026-02-21 18:16:52 -05:00 by yindo · 1 comment
yindo commented 2026-02-21 18:16:52 -05:00
Owner
Copy Link
Copy Source

Originally created by @BingranLIU on GitHub (Nov 11, 2024).

Self Checks

  • This is only for bug report, if you would like to ask a question, 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).
  • [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.

Dify version

0.11.0

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

In Dify versions higher than 0.9.0, when utilizing the agent module, if the prompt exceeds 1000 characters, the GLM-4 series models will lose the capability to retain chat history.
1.I input a 1200-character Chinese prompt. The first input was: “Please remember my name.” The second input was: “What is my name?” The model provided an incorrect response.
2.I conducted a comparative test with a 900-character Chinese prompt. The first input was: “Please remember my name.” The second input was: “What is my name?” The model correctly answered my question.

✔️ Expected Behavior

When the prompt exceeds 1000 Chinese characters, Dify agent module’s invocation of the GLM-4 series models still maintains the ability to understand chat-history context.

❌ Actual Behavior

For prompts exceeding 1000 Chinese characters, the GLM-4-ChatBot constructed by the “agent” is unable to maintain chat-history context.

Originally created by @BingranLIU on GitHub (Nov 11, 2024). ### Self Checks - [X] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [X] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [X] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [X] [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:） - [X] Please do not modify this template :) and fill in all the required fields. ### Dify version 0.11.0 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce In Dify versions higher than 0.9.0, when utilizing the agent module, if the prompt exceeds 1000 characters, the GLM-4 series models will lose the capability to retain chat history. 1.I input a 1200-character Chinese prompt. The first input was: “Please remember my name.” The second input was: “What is my name?” The model provided an incorrect response. 2.I conducted a comparative test with a 900-character Chinese prompt. The first input was: “Please remember my name.” The second input was: “What is my name?” The model correctly answered my question. ### ✔️ Expected Behavior When the prompt exceeds 1000 Chinese characters, Dify agent module’s invocation of the GLM-4 series models still maintains the ability to understand chat-history context. ### ❌ Actual Behavior For prompts exceeding 1000 Chinese characters, the GLM-4-ChatBot constructed by the “agent” is unable to maintain chat-history context.
yindo added the 🐞 bug label 2026-02-21 18:16:52 -05:00
yindo closed this issue 2026-02-21 18:16:52 -05:00
yindo commented 2026-02-21 18:16:53 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Nov 11, 2024):

@Ranyi176 Hello there! I'm here to help you with any bugs, questions, or contributions while you wait for a human maintainer. I'm looking forward to assisting you!

The issue you're experiencing with the GLM-4 series models losing chat history retention when prompts exceed 1000 characters might be related to how Dify manages input prompts and chat history. Dify uses a class called SimplePromptTransform to handle prompt transformations for both chat and completion models. This class ensures that the total token count does not exceed the model's limit by managing chat history with a TokenBufferMemory [1].

In your case, it seems that when the prompt length exceeds a certain threshold, the chat history might not be retained correctly. You might want to check the configuration settings for token limits in the glm_4_long.yaml file, where the max_tokens parameter can be customized. The default value is set to 1024, with a maximum of 4095 [2]. Adjusting this setting might help in retaining chat history for longer prompts.

To continue talking to Dosu, mention @dosu.

@dosubot[bot] commented on GitHub (Nov 11, 2024): <!-- Greeting --> @Ranyi176 Hello there! I'm here to help you with any bugs, questions, or contributions while you wait for a human maintainer. I'm looking forward to assisting you! <!-- Answer --> The issue you're experiencing with the GLM-4 series models losing chat history retention when prompts exceed 1000 characters might be related to how Dify manages input prompts and chat history. Dify uses a class called `SimplePromptTransform` to handle prompt transformations for both chat and completion models. This class ensures that the total token count does not exceed the model's limit by managing chat history with a `TokenBufferMemory` <sup>[[1]](https://github.com/langgenius/dify/blob/main/api/core/prompt/simple_prompt_transform.py)</sup>. In your case, it seems that when the prompt length exceeds a certain threshold, the chat history might not be retained correctly. You might want to check the configuration settings for token limits in the `glm_4_long.yaml` file, where the `max_tokens` parameter can be customized. The default value is set to 1024, with a maximum of 4095 <sup>[[2]](https://github.com/langgenius/dify/blob/main/api/core/model_runtime/model_providers/zhipuai/llm/glm_4_long.yaml)</sup>. Adjusting this setting might help in retaining chat history for longer prompts. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.*
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Knowledge Pipeline
Proposal
WIP
cloud
dependencies
feat:webapp
good first issue
outdated
performance
pull-request
Mirrored from GitHub Pull Request
 python:uv
refactor
size:L
stale
status: accepting prs
status: blocked
status: in discussion
stt & tts
vulnerablity
web
wont-fix
🌊 feat:workflow
🌚 invalid
🐞 bug
👻 feat:rag
💪 enhancement
📚 documentation
🔨 feat:tools
🙋‍♂️ question
🙏 help wanted
🤔 cant-reproduce
🤖 feat:agent
No Label 🐞 bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#6693
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?