soichisumi
/

qwen3-reranker-0.6b-mlx

Model card Files Files and versions

Qwen3-Reranker-0.6B — MLX (bf16)

Qwen/Qwen3-Reranker-0.6B の MLX 変換 (bfloat16、量子化なし)。1.1 GB。

変換: mlx_lm convert --hf-path Qwen/Qwen3-Reranker-0.6B --mlx-path . --dtype bfloat16
評価: mteb/scidocs-reranking 30 クエリ / 897 ペア
結果: Kendall τ = 0.9769, nDCG@10 Δ = +0.0033 → PASS+

Downloads last month: 15

Safetensors

Model size

0.6B params

Tensor type

BF16

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for soichisumi/qwen3-reranker-0.6b-mlx

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-Reranker-0.6B

Finetuned

(16)

this model