Embeddings and rerankers · Alibaba Qwen

Qwen3 Reranker 0.6B

Hugging Face: Qwen/Qwen3-Reranker-0.6B · checked 2026-10-13

Cross-encoder reranker to put after any embedder.

Datasheet

Maker	Alibaba Qwen
Parameters	596 MVerifiedsafetensors.total from the HF API Source
Licence	Apache-2.0Verified
Access	Public download
Task (HF pipeline)	text-ranking
First published on HF	2025-05-29
Last updated on HF	2026-04-16
Weights in memory	1.2 GB at 16-bit · 0.4 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build.

Benchmark

65.8

MTEB-R (reranking) · #1 of 2 scored in this category

MTEB-R, model card table (reranker metric; not comparable with embedding scores). Source

See the full embeddings and rerankers ranking

Used in

Offline document search with RAG
A medical literature helper (not medical advice)

The description summarises the model card at https://huggingface.co/Qwen/Qwen3-Reranker-0.6B. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.