langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-08-27 11:11:21 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

Knowledge Retrieval Token Limit Clarification #6006

New Issue
Closed
opened 2026-02-21 18:13:38 -05:00 by yindo · 3 comments
yindo commented 2026-02-21 18:13:38 -05:00
Owner
Copy Link
Copy Source

Originally created by @NourKhdour on GitHub (Oct 10, 2024).

Originally assigned to: @Yawen-1010, @AllenWriter on GitHub.

Self Checks

  • I have searched for existing issues search for existing issues, including closed ones.
  • I confirm that I am using English to submit 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.

Provide a description of requested docs changes

I would like clarification and additional documentation regarding the maximum token length for Knowledge Retrieval queries, Is it the same as Retrieval Testing up to 200 characters or it follows the same constraints as the model’s token limit?

Questions:
Is it better to pass a long document as one chunk or split it into smaller pieces for more effective retrieval?
If splitting is required, what is the optimal chunk size (in tokens or characters) for accurate and complete retrieval?

Originally created by @NourKhdour on GitHub (Oct 10, 2024). Originally assigned to: @Yawen-1010, @AllenWriter on GitHub. ### Self Checks - [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 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. ### Provide a description of requested docs changes I would like clarification and additional documentation regarding the maximum token length for Knowledge Retrieval queries, Is it the same as Retrieval Testing up to 200 characters or it follows the same constraints as the model’s token limit? **Questions:** Is it better to pass a long document as one chunk or split it into smaller pieces for more effective retrieval? If splitting is required, what is the optimal chunk size (in tokens or characters) for accurate and complete retrieval?
yindo added the 📚 documentation label 2026-02-21 18:13:38 -05:00
yindo closed this issue 2026-02-21 18:13:38 -05:00
yindo commented 2026-02-21 18:13:38 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Oct 10, 2024):

The maximum token length for Knowledge Retrieval queries in Dify isn't specified in the available information. However, for document splitting, it's recommended to split long documents into chunks with a maximum of 4000 tokens per segment. A chunk overlap of about 200 tokens is also suggested for better retrieval accuracy [1].

Splitting documents into smaller pieces is generally more effective for retrieval, as it ensures that each segment fits within the model's context window and improves semantic matching.

To continue talking to Dosu, mention @dosu.

@dosubot[bot] commented on GitHub (Oct 10, 2024): <!-- Answer --> The maximum token length for Knowledge Retrieval queries in Dify isn't specified in the available information. However, for document splitting, it's recommended to split long documents into chunks with a maximum of 4000 tokens per segment. A chunk overlap of about 200 tokens is also suggested for better retrieval accuracy <sup>[[1]](https://github.com/langgenius/dify-docs/blob/main/en/guides/knowledge-base/create-knowledge-and-upload-documents.md)</sup>. Splitting documents into smaller pieces is generally more effective for retrieval, as it ensures that each segment fits within the model's context window and improves semantic matching. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.*
yindo commented 2026-02-21 18:13:38 -05:00
Author
Owner
Copy Link
Copy Source

@Yawen-1010 commented on GitHub (Oct 11, 2024):

The maximum length for Knowledge Retrieval queries is the same as Retrieval Testing up to 200 characters. We will specify it in our product document. Thanks for raising this question. @NourKhdour

@Yawen-1010 commented on GitHub (Oct 11, 2024): The maximum length for Knowledge Retrieval queries is the same as Retrieval Testing up to 200 characters. We will specify it in our product document. Thanks for raising this question. @NourKhdour
yindo commented 2026-02-21 18:13:39 -05:00
Author
Owner
Copy Link
Copy Source

@Yawen-1010 commented on GitHub (Oct 12, 2024):

Document has been updated.
https://docs.dify.ai/guides/workflow/node/knowledge-retrieval

@Yawen-1010 commented on GitHub (Oct 12, 2024): Document has been updated. https://docs.dify.ai/guides/workflow/node/knowledge-retrieval
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 📚 documentation
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#6006
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?