Use of Image embedding model #3924

New Issue

2025-11-11T15:42:39-06:00

GiteaMirror commented

2025-11-11 15:42:39 -06:00

Originally created by @rpaGuyai on GitHub (Feb 18, 2025).

I have tried switching to different text to text embedding models and it works fine but when I try using any of image embedding model, for .e.g. openai/clip-vit-large-patch14, I get error while uploading document as "400: 'NoneType' object has no attribute 'encode'.

My goal is to get the model understand images within the documents upload in RAG and even for the documents uploaded from frontend.

So I ma trying to use an image embedder to achieve this and a vision LLM model.

Please do guide me on my approach as well if I am doing it incorrectly.

Thanks

Originally created by @rpaGuyai on GitHub (Feb 18, 2025). I have tried switching to different text to text embedding models and it works fine but when I try using any of image embedding model, for .e.g. openai/clip-vit-large-patch14, I get error while uploading document as "**400: 'NoneType' object has no attribute 'encode'**. My goal is to get the model understand images within the documents upload in RAG and even for the documents uploaded from frontend. So I ma trying to use an image embedder to achieve this and a vision LLM model. Please do guide me on my approach as well if I am doing it incorrectly. Thanks

GiteaMirror closed this issue

2025-11-11 15:42:39 -06:00

GiteaMirror referenced this issue

2026-04-19 20:10:37 -05:00

[GH-ISSUE #3924] How to revert back to the old version? #13433

GiteaMirror referenced this issue

2026-04-25 03:27:29 -05:00

[GH-ISSUE #3924] How to revert back to the old version? #28961

GiteaMirror referenced this issue

2026-05-05 13:16:51 -05:00

[GH-ISSUE #3924] How to revert back to the old version? #52099

GiteaMirror referenced this issue