Request: Add Support for Qwen3-Reranker Model

[gy9527]

Thanks for your amazing work!

[giladgd]

I've opened #506 and will release a new version in the next few days after running tests on various hardware and platforms

[giladgd]

@gy9527 I've just release version 3.14.0 with support for Qwen 3 Reranker.

The new Qwen3 Reranker support only works with recent models converted with llama.cpp release b6578 or later.
Here are models quantizations I made that you can use right away:

Model	URI	Size
Qwen3 Reranker 0.6B	hf:giladgd/Qwen3-Reranker-0.6B-GGUF:Q8_0	639.2MB
Qwen3 Reranker 4B	hf:giladgd/Qwen3-Reranker-4B-GGUF:Q4_K_M	2.5GB
Qwen3 Reranker 8B	hf:giladgd/Qwen3-Reranker-8B-GGUF:Q4_K_M	4.7GB