[GH-ISSUE #884] [BUG]: Use remote embedder fails #526

Closed
opened 2026-02-22 18:19:58 -05:00 by yindo · 3 comments
Owner

Originally created by @egSat on GitHub (Mar 11, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/884

How are you running AnythingLLM?

Docker (local)

What happened?

Hi all
I'm trying to use a remote embedder, loaded from text-generation-webui. I set thte ip address and port, and Max embedding chunk length to 384. I use LanceDB.
I removed the file and loaded again (as recommended when switching embedder), but it shows the following error from logs:
Adding new vectorized document into namespace emfpruebas
Chunks created from document: 13
Inserting vectorized chunks into LanceDB collection.
addDocumentToNamespace LanceDBError: Append with different schema: original=Field(id=0, name=vector, type=fixed_size_list:float:384)
Field(id=1, name=id, type=string)
Field(id=2, name=url, type=string)
Field(id=3, name=title, type=string)
Field(id=4, name=docAuthor, type=string)
Field(id=5, name=description, type=string)
Field(id=6, name=docSource, type=string)
Field(id=7, name=chunkSource, type=string)
Field(id=8, name=published, type=string)
Field(id=9, name=wordCount, type=double)
Field(id=10, name=token_count_estimate, type=double)
Field(id=11, name=text, type=string)
new=Field(id=0, name=vector, type=fixed_size_list:float:768)
Field(id=1, name=id, type=string)
Field(id=2, name=url, type=string)
Field(id=3, name=title, type=string)
Field(id=4, name=docAuthor, type=string)
Field(id=5, name=description, type=string)
Field(id=6, name=docSource, type=string)
Field(id=7, name=chunkSource, type=string)
Field(id=8, name=published, type=string)
Field(id=9, name=wordCount, type=double)
Field(id=10, name=token_count_estimate, type=double)
Field(id=11, name=text, type=string)

Failed to vectorize file with stuff.docx

Are there known steps to reproduce?

Use a remote embedder with Max embedding chunk length different than 768

Originally created by @egSat on GitHub (Mar 11, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/884 ### How are you running AnythingLLM? Docker (local) ### What happened? Hi all I'm trying to use a remote embedder, loaded from text-generation-webui. I set thte ip address and port, and Max embedding chunk length to 384. I use LanceDB. I removed the file and loaded again (as recommended when switching embedder), but it shows the following error from logs: Adding new vectorized document into namespace emfpruebas Chunks created from document: 13 Inserting vectorized chunks into LanceDB collection. addDocumentToNamespace LanceDBError: Append with different schema: original=Field(id=0, name=vector, type=fixed_size_list:float:384) Field(id=1, name=id, type=string) Field(id=2, name=url, type=string) Field(id=3, name=title, type=string) Field(id=4, name=docAuthor, type=string) Field(id=5, name=description, type=string) Field(id=6, name=docSource, type=string) Field(id=7, name=chunkSource, type=string) Field(id=8, name=published, type=string) Field(id=9, name=wordCount, type=double) Field(id=10, name=token_count_estimate, type=double) Field(id=11, name=text, type=string) new=Field(id=0, name=vector, type=fixed_size_list:float:768) Field(id=1, name=id, type=string) Field(id=2, name=url, type=string) Field(id=3, name=title, type=string) Field(id=4, name=docAuthor, type=string) Field(id=5, name=description, type=string) Field(id=6, name=docSource, type=string) Field(id=7, name=chunkSource, type=string) Field(id=8, name=published, type=string) Field(id=9, name=wordCount, type=double) Field(id=10, name=token_count_estimate, type=double) Field(id=11, name=text, type=string) Failed to vectorize file with stuff.docx ### Are there known steps to reproduce? Use a remote embedder with Max embedding chunk length different than 768
yindo added the possible bug label 2026-02-22 18:19:58 -05:00
yindo closed this issue 2026-02-22 18:19:58 -05:00
Author
Owner

@egSat commented on GitHub (Mar 11, 2024):

Hi,

removing the data from folder (lancedb/workspace-lance/data) fixes it.
I don't know just removing the document from web should be sufficient, so I leave this issue open.

thank you very much for this amazing project

@egSat commented on GitHub (Mar 11, 2024): Hi, removing the data from folder (lancedb/workspace-lance/data) fixes it. I don't know just removing the document from web should be sufficient, so I leave this issue open. thank you very much for this amazing project
Author
Owner

@timothycarambat commented on GitHub (Mar 11, 2024):

Do you know which tag of docker you are/were on? This was an old bug we had a long time ago that was due to lanceDB package that we bumped and fixed.

If you dont know tag the value for vectordb in the /app/server/package.json file in the container would be nice to have. It should be 0.1.19

@timothycarambat commented on GitHub (Mar 11, 2024): Do you know which tag of docker you are/were on? This was an old bug we had a long time ago that was due to lanceDB package that we bumped and fixed. If you dont know tag the value for `vectordb` in the `/app/server/package.json` file in the container would be nice to have. It should be `0.1.19`
Author
Owner

@egSat commented on GitHub (Mar 11, 2024):

Yes, it shows 0.1.19 for vectorDB. I pulled the image two days ago so I assume it's the latest one

@egSat commented on GitHub (Mar 11, 2024): Yes, it shows 0.1.19 for vectorDB. I pulled the image two days ago so I assume it's the latest one
yindo changed title from [BUG]: Use remote embedder fails to [GH-ISSUE #884] [BUG]: Use remote embedder fails 2026-06-05 14:35:48 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#526