Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 22:23:50 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #1671] [BUG]: OpenAI does not have access to text-moderation-007 #1088

New Issue
Closed
opened 2026-02-22 18:23:04 -05:00 by yindo · 11 comments
yindo commented 2026-02-22 18:23:04 -05:00
Owner
Copy Link
Copy Source

Originally created by @IMTanuki on GitHub (Jun 12, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1671

Originally assigned to: @timothycarambat on GitHub.

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Installed a default OpenAI setup:
LLM Provider - OpenAI
workspace chat model - gpt-4o
-- also tried gpt-4, gpt-0613
chat mode - chat
history - 20
prompt - default
refusal response - default
temp - .7

Prompt:

  • can be anything
  • I tried a complex request, as well as 'Hello'

Error:
OpenAI::CreateModeration failed with: 403 Project proj_xxx does not have access to model text-moderation-007

Notes

  • text-moderation-007 is not a listed model in OpenAi's project models
  • I understand this is a dedicated moderation endpoint to detect inappropriate requests
  • this error is listed in OpenAI community (as of 1 week ago), but no answer is listed

Is there any way to suppress this call?

Are there known steps to reproduce?

See notes above.
Error occurs regardless of the prompt.

Originally created by @IMTanuki on GitHub (Jun 12, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1671 Originally assigned to: @timothycarambat on GitHub. ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? **Installed a default OpenAI setup:** LLM Provider - OpenAI workspace chat model - gpt-4o -- also tried gpt-4, gpt-0613 chat mode - chat history - 20 prompt - default refusal response - default temp - .7 **Prompt:** - can be anything - I tried a complex request, as well as 'Hello' **Error:** OpenAI::CreateModeration failed with: 403 Project `proj_xxx` does not have access to model `text-moderation-007` **Notes** - text-moderation-007 is not a listed model in OpenAi's project models - I understand this is a dedicated moderation endpoint to detect inappropriate requests - this error is listed in OpenAI community (as of 1 week ago), but no answer is listed Is there any way to suppress this call? ### Are there known steps to reproduce? See notes above. Error occurs regardless of the prompt.
yindo added the possible bug label 2026-02-22 18:23:04 -05:00
yindo closed this issue 2026-02-22 18:23:04 -05:00
yindo commented 2026-02-22 18:23:04 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 13, 2024):

Referring to this discussion, correct?
https://community.openai.com/t/no-access-to-text-moderation-models/760242/12

@timothycarambat commented on GitHub (Jun 13, 2024): Referring to this discussion, correct? https://community.openai.com/t/no-access-to-text-moderation-models/760242/12
yindo commented 2026-02-22 18:23:04 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 13, 2024):

yup
tbh - my goal w/ anything is to migrate away from openai and it's progressively limited functionality, including text moderation. Using OpenAI is step 1 in testing anything and would simplify my initial setup.

  • so, is there a fix for this when using OpenAI?

I've got a new MSI w/ 64GB and and RTX 4090 w 16GB and I'm hoping that will drive an LLM from Mistral. I guess my followup question would be:

  • does anything use a variant of text-moderation w/ other open source LLM's?

Thx!

@IMTanuki commented on GitHub (Jun 13, 2024): yup tbh - my goal w/ anything is to migrate away from openai and it's progressively limited functionality, including text moderation. Using OpenAI is step 1 in testing anything and would simplify my initial setup. - so, is there a fix for this when using OpenAI? I've got a new MSI w/ 64GB and and RTX 4090 w 16GB and I'm hoping that will drive an LLM from Mistral. I guess my followup question would be: - does anything use a variant of text-moderation w/ other open source LLM's? Thx!
yindo commented 2026-02-22 18:23:04 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 13, 2024):

does anything use a variant of text-moderation w/ other open-source LLM's?

Only OpenAi has moderation on chats and its high time we just remove it/make it opt-in. It originally existed simply because we didn't want people accidentally getting themselves banned but to force it as enabled, but only OpenAI, makes little sense now since all others are unmoderated and at the end of the day...its your key!

@timothycarambat commented on GitHub (Jun 13, 2024): > does anything use a variant of text-moderation w/ other open-source LLM's? Only OpenAi has moderation on chats and its high time we just remove it/make it opt-in. It originally existed simply because we didn't want people accidentally getting themselves banned but to force it as enabled, but **only** OpenAI, makes little sense now since all others are unmoderated and at the end of the day...its your key!
yindo commented 2026-02-22 18:23:04 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 13, 2024):

Can I ask - are you with OpenAI or Mintplex?
Either way, 100 agree it's unnecessary. I have a quirky sense of humor and in chatGPT, I keep getting warnings about inappropriate prompts - like, I'm chatting with software...who will get offended?

Just to confirm - with all other LLM's, no issues, right?

@IMTanuki commented on GitHub (Jun 13, 2024): Can I ask - are you with OpenAI or Mintplex? Either way, 100 agree it's unnecessary. I have a quirky sense of humor and in chatGPT, I keep getting warnings about inappropriate prompts - like, I'm chatting with software...who will get offended? Just to confirm - with all other LLM's, no issues, right?
yindo commented 2026-02-22 18:23:04 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 13, 2024):

I am with the core-team for Mintplex Labs which builds this tool.

Just to confirm - with all other LLM's, no issues, right?

Correct! If you get any refusals it's either from the model directly or if using some cloud-based LLM they may do some censoring, but nothing we control.

@timothycarambat commented on GitHub (Jun 13, 2024): I am with the core-team for Mintplex Labs which builds this tool. > Just to confirm - with all other LLM's, no issues, right? Correct! If you get any refusals it's either from the model directly or if using some cloud-based LLM they may do some censoring, but nothing we control.
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 13, 2024):

Cool! Thx for the clarification. Just downloaded anything (Win app) last night - UI and options look great! Hope to get it running w/ a Mistral LLM later today.

@IMTanuki commented on GitHub (Jun 13, 2024): Cool! Thx for the clarification. Just downloaded anything (Win app) last night - UI and options look great! Hope to get it running w/ a Mistral LLM later today.
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 14, 2024):

Sorry for re-opening this issue but I was wondering - I have my own (very basic) openAI client and I never get the text-moderation-007 error. Is there sth unique about how anything interfaces with openAI?

@IMTanuki commented on GitHub (Jun 14, 2024): Sorry for re-opening this issue but I was wondering - I have my own (very basic) openAI client and I never get the text-moderation-007 error. Is there sth unique about how anything interfaces with openAI?
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 14, 2024):

If you are not using the moderation endpoint, then you would never see this error

@timothycarambat commented on GitHub (Jun 14, 2024): If you are not using the moderation endpoint, then you would never see this error
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 14, 2024):

Not to beat this to death, but if my client doesn't encounter the text-moderation-007 error, presumably bc I'm not using the moderation endpoint, then can't Mintplex likewise remove its use of the same endpoint? Unless I'm missing something, it seems like that should be a relatively easy solution to this issue.

@IMTanuki commented on GitHub (Jun 14, 2024): Not to beat this to death, but if my client doesn't encounter the text-moderation-007 error, presumably bc I'm not using the moderation endpoint, then can't Mintplex likewise remove its use of the same endpoint? Unless I'm missing something, it seems like that should be a relatively easy solution to this issue.
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 14, 2024):

That is what I was saying in https://github.com/Mintplex-Labs/anything-llm/issues/1671#issuecomment-2166116264

@timothycarambat commented on GitHub (Jun 14, 2024): That is what I was saying in https://github.com/Mintplex-Labs/anything-llm/issues/1671#issuecomment-2166116264
yindo commented 2026-02-22 18:23:05 -05:00
Author
Owner
Copy Link
Copy Source

@IMTanuki commented on GitHub (Jun 17, 2024):

Per the note in OpenAI forum, using an api key from the default project seems to work.

@IMTanuki commented on GitHub (Jun 17, 2024): Per the note in OpenAI forum, using an api key from the default project seems to work.
yindo changed title from [BUG]: OpenAI does not have access to text-moderation-007 to [GH-ISSUE #1671] [BUG]: OpenAI does not have access to text-moderation-007 2026-06-05 14:38:55 -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#1088
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?