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 #3327] [BUG]: embedding with OpenAI API, 429 Rate limit #2141

New Issue
Closed
opened 2026-02-22 18:28:19 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:28:19 -05:00
Owner
Copy Link
Copy Source

Originally created by @chourete on GitHub (Feb 23, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3327

How are you running AnythingLLM?

AnythingLLM Desktop app

What happened?

Hi Developer

I’m encountering an issue when uploading a large .txt file (29MB). The system returns the following error:

OpenAI Failed to embed: [failed_to_embed]: 429 Rate limit reached for text-embedding-3-small in organization org-XXXXX on tokens per min (TPM): Limit 1,000,000, Used 1,000,000, Requested 38,761. Please try again in 2.325s.

It seems that the embedding process is exceeding the rate limit. Should we implement a rate limit delay or batch processing to avoid hitting the TPM restriction?

Are there known steps to reproduce?

Embedding Provider, both

Image

OR

Image

Originally created by @chourete on GitHub (Feb 23, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3327 ### How are you running AnythingLLM? AnythingLLM Desktop app ### What happened? Hi Developer I’m encountering an issue when uploading a large .txt file (29MB). The system returns the following error: ``` OpenAI Failed to embed: [failed_to_embed]: 429 Rate limit reached for text-embedding-3-small in organization org-XXXXX on tokens per min (TPM): Limit 1,000,000, Used 1,000,000, Requested 38,761. Please try again in 2.325s. ``` It seems that the embedding process is exceeding the rate limit. Should we implement a rate limit delay or batch processing to avoid hitting the TPM restriction? ### Are there known steps to reproduce? Embedding Provider, both ![Image](https://github.com/user-attachments/assets/c01f9694-04a4-4898-8891-c96005fce0ca) OR ![Image](https://github.com/user-attachments/assets/c20ab8d3-84e7-440c-aa3b-216c7d41ae89)
yindo added the possible bug label 2026-02-22 18:28:19 -05:00
yindo closed this issue 2026-02-22 18:28:19 -05:00
yindo commented 2026-02-22 18:28:19 -05:00
Author
Owner
Copy Link
Copy Source

@shatfield4 commented on GitHub (Feb 24, 2025):

We do have batching already in the OpenAI provider to prevent rate limiting from happening. If you're using the generic OpenAI provider try reducing the max concurrent chunks to even less. I also just want to let you know that uploading a .txt. file with 29 MB of text is not going to give the results you are probably looking for when using RAG. It will be very hard for RAG to find the data you are looking for and will not give accurate results. Closing for now.

@shatfield4 commented on GitHub (Feb 24, 2025): We do have batching already in the OpenAI provider to prevent rate limiting from happening. If you're using the generic OpenAI provider try reducing the max concurrent chunks to even less. I also just want to let you know that uploading a `.txt.` file with 29 MB of text is not going to give the results you are probably looking for when using RAG. It will be very hard for RAG to find the data you are looking for and will not give accurate results. Closing for now.
yindo commented 2026-02-22 18:28:19 -05:00
Author
Owner
Copy Link
Copy Source

@chourete commented on GitHub (Feb 24, 2025):

Hi developer, this is just a test, for case in point, if multiple users are uploading files at the same time, it can easily lead to a rate limit.
For example, when librechat uses embedding with openai, it has a retry delay seconds in the backend(rag-api). I think this is a good idea...

for anythingllm
Even if it's a text-only 29MB, or 14MB, I've found that he triggers an error causing the file to fail to be embedded.

@chourete commented on GitHub (Feb 24, 2025): Hi developer, this is just a test, for case in point, if multiple users are uploading files at the same time, it can easily lead to a rate limit. For example, when librechat uses embedding with openai, it has a retry delay seconds in the backend(rag-api). I think this is a good idea... for anythingllm Even if it's a text-only 29MB, or 14MB, I've found that he triggers an error causing the file to fail to be embedded.
yindo changed title from [BUG]: embedding with OpenAI API, 429 Rate limit to [GH-ISSUE #3327] [BUG]: embedding with OpenAI API, 429 Rate limit 2026-06-05 14:44:47 -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#2141
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?