support Qwen3 embedding

[bino282]

[BobLiu20]

Qwen3 embedding is ok. But not qwen3 reranker

[wirthual]

Try an adapted version like this:

infinity_emb v2 --model-id tomaarsen/Qwen3-Reranker-0.6B-seq-cls

Note: You might be required to update your transformers version.

[Torala]

With Transformers versions earlier than 4.51.0, you may encounter the following error: KeyError: 'qwen3'

[fanyang89]

I run qwen3-reranker with the following commands:

# Thanks @Torala 
uv add "transformers>=4.51.0"

# https://github.com/huggingface/optimum/issues/2277
sed -i 's/raise RuntimeError/print/g' $(fd -HI __init__.py | grep 'bettertransformer/_')

# Thanks @wirthual
uv run infinity_emb v2 --model-id "tomaarsen/Qwen3-Reranker-0.6B-seq-cls"

[dotmobo]

+1 for reranker