langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-21 12:05:30 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

After updating to 0.9.2, workflow with knowledge retrieval became unexecutable due to lack of rerank model #6089

New Issue
Closed
opened 2026-02-21 18:14:02 -05:00 by yindo · 10 comments
yindo commented 2026-02-21 18:14:02 -05:00
Owner
Copy Link
Copy Source

Originally created by @mabuchs on GitHub (Oct 14, 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

0.9.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

  1. In model setup, only setup LLM models. (without any rerank models)
  2. Create workflow and include knowledge retrieval node
  3. Try [Run]

DSL for the workflow is attatched.
sampleflow.yml.txt

✔️ Expected Behavior

Workflow can be executed even if rerank model is not enabled on Dify.

❌ Actual Behavior

After reproduction steps, it shows error "Rerank model is required". This didn't occur until v0.9.1.
image

Originally created by @mabuchs on GitHub (Oct 14, 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)). - [ ] [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 0.9.2 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce 1. In model setup, only setup LLM models. (without any rerank models) 2. Create workflow and include knowledge retrieval node 3. Try \[Run\] DSL for the workflow is attatched. [sampleflow.yml.txt](https://github.com/user-attachments/files/17366557/sampleflow.yml.txt) ### ✔️ Expected Behavior Workflow can be executed even if rerank model is not enabled on Dify. ### ❌ Actual Behavior After reproduction steps, it shows error "Rerank model is required". This didn't occur until v0.9.1. <img width="1283" alt="image" src="https://github.com/user-attachments/assets/0cda3320-535f-4bbd-8e2e-dcb18c673c3e">
yindo added the 🐞 bug label 2026-02-21 18:14:02 -05:00
yindo closed this issue 2026-02-21 18:14:02 -05:00
yindo commented 2026-02-21 18:14:02 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Oct 14, 2024):

Hey @mabuchs! 👋 I'm here to help you with any bugs, questions, or contributions you have for Dify. I'm an expert bot and can assist you while you wait for a human maintainer.

In Dify 0.9.2, the workflow with knowledge retrieval requires a rerank model due to the use of hybrid search techniques that improve recall by combining different retrieval technologies. The rerank model is necessary to merge and normalize query results, ensuring the most relevant content is identified [1][2].

To bypass the rerank model requirement, you can use the Weighted Score setting. This setting uses internal scoring mechanisms without needing an external rerank model, allowing you to adjust the weight ratio sliders for semantics or keywords to select the most appropriate content matching strategy [3].

To continue talking to Dosu, mention @dosu.

@dosubot[bot] commented on GitHub (Oct 14, 2024): <!-- Greeting --> Hey @mabuchs! 👋 I'm here to help you with any bugs, questions, or contributions you have for Dify. I'm an expert bot and can assist you while you wait for a human maintainer. <!-- Answer --> In Dify 0.9.2, the workflow with knowledge retrieval requires a rerank model due to the use of hybrid search techniques that improve recall by combining different retrieval technologies. The rerank model is necessary to merge and normalize query results, ensuring the most relevant content is identified <sup>[[1]](https://github.com/langgenius/dify-docs/blob/main/en/guides/workflow/node/knowledge-retrieval.md)</sup><sup>[[2]](https://github.com/langgenius/dify/blob/main/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py)</sup>. To bypass the rerank model requirement, you can use the **Weighted Score** setting. This setting uses internal scoring mechanisms without needing an external rerank model, allowing you to adjust the weight ratio sliders for semantics or keywords to select the most appropriate content matching strategy <sup>[[3]](https://github.com/langgenius/dify-docs/blob/main/en/guides/knowledge-base/integrate-knowledge-within-application.md)</sup>. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.*
yindo commented 2026-02-21 18:14:02 -05:00
Author
Owner
Copy Link
Copy Source

@crazywoola commented on GitHub (Oct 14, 2024):

Set a rerank model here.

image
@crazywoola commented on GitHub (Oct 14, 2024): Set a rerank model here. <img width="1000" alt="image" src="https://github.com/user-attachments/assets/04f9609a-82f2-4bbe-b1a3-3e7d1d23185c">
yindo commented 2026-02-21 18:14:02 -05:00
Author
Owner
Copy Link
Copy Source

@mabuchs commented on GitHub (Oct 14, 2024):

Thank you for response.

Yes, I know how to set rerank model. But I want to know if I can use knowledge retrieval node in an environment without rerank model (due to restirction of model selection in company).

Let me confirm: using knowledge retrieval without rerank model became completely unsupported (even if using Weighted Score)?
(In v0.9.1, I could use it if I use only one knowledge and weighted score)

@mabuchs commented on GitHub (Oct 14, 2024): Thank you for response. Yes, I know how to set rerank model. But I want to know if I can use knowledge retrieval node in an environment without rerank model (due to restirction of model selection in company). Let me confirm: using knowledge retrieval without rerank model became completely unsupported (even if using Weighted Score)? (In v0.9.1, I could use it if I use only one knowledge and weighted score)
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@mabuchs commented on GitHub (Oct 15, 2024):

@crazywoola Could you kindly confirm the question above?

@mabuchs commented on GitHub (Oct 15, 2024): @crazywoola Could you kindly confirm the question above?
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@fniu commented on GitHub (Oct 20, 2024):

This issue remains. I cannot run my app from the workflow editor due to this error, while the app can be published without any problem. @crazywoola Could you please reconsider this issue? I have also sent an email for support.

@fniu commented on GitHub (Oct 20, 2024): This issue remains. I cannot run my app from the workflow editor due to this error, while the app can be published without any problem. @crazywoola Could you please reconsider this issue? I have also sent an email for support.
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@kurokobo commented on GitHub (Oct 20, 2024):

@mabuchs @fniu
Hi, could you please change the setting of the knowledge retrieval node back to Weighted Score again and check if it works as it did before?
image

@kurokobo commented on GitHub (Oct 20, 2024): @mabuchs @fniu Hi, could you please change the setting of the knowledge retrieval node back to Weighted Score again and check if it works as it did before? ![image](https://github.com/user-attachments/assets/a38e9be0-a7ff-4774-8feb-2fd3d7eac4ee)
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@fniu commented on GitHub (Oct 20, 2024):

@mabuchs @fniu Hi, could you please change the setting of the knowledge retrieval node back to Weighted Score again and check if it works as it did before?

I think this issue may only exist for workflows that are created before 0.9.2. I have tested it with new workflows which seem to work fine. For existing workflows, I have changed different settings back and forth, and the issue still remains. One action may be able to fix it though (I came across by accident), which is to click tab "Monitoring" and then back to "Orchestrate" and the error was gone.

@fniu commented on GitHub (Oct 20, 2024): > @mabuchs @fniu Hi, could you please change the setting of the knowledge retrieval node back to Weighted Score again and check if it works as it did before? I think this issue may only exist for workflows that are created before 0.9.2. I have tested it with new workflows which seem to work fine. For existing workflows, I have changed different settings back and forth, and the issue still remains. One action may be able to fix it though (I came across by accident), which is to click tab "Monitoring" and then back to "Orchestrate" and the error was gone.
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@kkjinping commented on GitHub (Oct 21, 2024):

@crazywoola @kurokobo There is still this problem in the external knowledge

@kkjinping commented on GitHub (Oct 21, 2024): @crazywoola @kurokobo There is still this problem in the external knowledge
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@MakabakaHuba commented on GitHub (Oct 26, 2024):

seem problem we meet😭

@MakabakaHuba commented on GitHub (Oct 26, 2024): seem problem we meet😭
yindo commented 2026-02-21 18:14:03 -05:00
Author
Owner
Copy Link
Copy Source

@mabuchs commented on GitHub (Oct 30, 2024):

I tried creating new workflow on latest version, then I confirmed this has been resolved. thank you!

@mabuchs commented on GitHub (Oct 30, 2024): I tried creating new workflow on latest version, then I confirmed this has been resolved. thank you!
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#6089
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?