Skip to content
This repository has been archived by the owner on Sep 18, 2024. It is now read-only.

Onnx weights for embedding models #770

Open
1951FDG opened this issue Jan 11, 2024 · 0 comments
Open

Onnx weights for embedding models #770

1951FDG opened this issue Jan 11, 2024 · 0 comments

Comments

@1951FDG
Copy link

1951FDG commented Jan 11, 2024

Hello,

I hope this is the right place to ask, it would be nice to have the onnx weights added for the models specified below, just like was done for:
https://huggingface.co/jinaai/jina-embeddings-v2-base-en/tree/main

models:
jina-embedding-t-en-v1
jina-embedding-s-en-v1
jina-embedding-b-en-v1
jina-embedding-l-en-v1

But then putting the weights both quantized and not quantized in a folder named onnx just like:
https://huggingface.co/Xenova/jina-embeddings-v2-base-en

onnx
- model.onnx
- model_quantized.onnx

I will try to convert one myself (jina-embedding-s-en-v1) soon to see if it works, but it would be nice to have official weights to compare results.

# for free to subscribe to this conversation on GitHub. Already have an account? #.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant