Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2401] [BUG]: Embedding fails with large documents #1559

New Issue
Closed
opened 2026-02-22 18:25:24 -05:00 by yindo · 1 comment
yindo commented 2026-02-22 18:25:24 -05:00
Owner
Copy Link
Copy Source

Originally created by @EDLLT on GitHub (Oct 1, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2401

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Embedding a large document fails no matter what document it is(The document in this case was 70MB)
I had to split up the document into 20MB files for it to embed properly.
I tested the an Ollama stella-en embedder, LM Studio mxbai embed large and AnythingLLM's native embedder all with the same issue

The weird thing is, after the embedding had failed, the vector db has a vector count above 0 despite no documents being cached/embedded whatsoever
image

I am running this on Ubuntu 24.04 by the way

Are there known steps to reproduce?

Get a large file(70MB in my case)
Select AnythingLLM's native embedder
Attempt to embed it then wait for it to fail

Originally created by @EDLLT on GitHub (Oct 1, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2401 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Embedding a large document fails no matter what document it is(The document in this case was 70MB) I had to split up the document into 20MB files for it to embed properly. I tested the an Ollama stella-en embedder, LM Studio mxbai embed large and AnythingLLM's native embedder all with the same issue The weird thing is, after the embedding had failed, the vector db has a vector count above 0 despite no documents being cached/embedded whatsoever ![image](https://github.com/user-attachments/assets/c5626322-a7f5-4757-8737-8688dc574aaa) I am running this on Ubuntu 24.04 by the way ### Are there known steps to reproduce? Get a large file(70MB in my case) Select AnythingLLM's native embedder Attempt to embed it then wait for it to fail
yindo added the possible bug label 2026-02-22 18:25:24 -05:00
yindo closed this issue 2026-02-22 18:25:24 -05:00
yindo commented 2026-02-22 18:25:24 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Oct 1, 2024):

This is a machine limitation - especially if you get this result with other providers. It is not described what you machine specs are but I am assuming CPU embedding is occurring. Likely OOM for trying to process so many text chunks at once. Additionally, if your context window for the selected embedding model is incorrect (for Ollama embedding) this will also fail because every embedding model has a defined limit of how many tokens it can vectorized in a given run - people just put in any value to that field and think it will work. Embedding models are just like LLM's and have token limits which can be found via the model card on HuggingFace or Ollama.com's model page

@timothycarambat commented on GitHub (Oct 1, 2024): This is a machine limitation - especially if you get this result with other providers. It is not described what you machine specs are but I am assuming CPU embedding is occurring. Likely OOM for trying to process so many text chunks at once. Additionally, if your context window for the selected embedding model is incorrect (for Ollama embedding) this will also fail because every embedding model has a defined limit of how many tokens it can vectorized in a given run - people just put in any value to that field and think it will work. Embedding models are just like LLM's and have token limits which can be found via the model card on HuggingFace or Ollama.com's model page
yindo changed title from [BUG]: Embedding fails with large documents to [GH-ISSUE #2401] [BUG]: Embedding fails with large documents 2026-06-05 14:41:25 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label possible 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: Mintplex-Labs/anything-llm#1559
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?