Skip to page content
SAVRN
Search Plan a project

Open-weight model · Text ranking

RK182X-EMBEDDING-Qwen3-Reranker-0.6B

by RKNNAI RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B

RK182X-EMBEDDING-Qwen3-Reranker-0.6B is an open-weight model for text ranking from RKNNAI, released under Apache License 2.0. Its published files total 597.8 MB.

本仓库提供由 Qwen/Qwen3-Reranker-0.6B 转换的 RKNN 模型。 - Model ID:RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B - 模型显示名称:RK182X-EMBEDDING-Qwen3-Reranker-0.6B - 源模型:Qwen/Qwen3-Reranker-0.6B - 模型类型:EMBEDDING ModelScope 完整下载: Hugging Face 完整下载: ModelScope 指定配置下载: Hugging…

Parameters—
Context—
Weights316.5 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—

Model Card

By RKNNAI, published under apache-2.0, revision ffe4f37aaffe.

本仓库提供由 Qwen/Qwen3-Reranker-0.6B 转换的 RKNN 模型。 - Model ID:RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B - 模型显示名称:RK182X-EMBEDDING-Qwen3-Reranker-0.6B - 源模型:Qwen/Qwen3-Reranker-0.6B - 模型类型:EMBEDDING ModelScope 完整下载: Hugging Face 完整下载: ModelScope 指定配置下载: Hugging Face 指定配置下载: - 使用配套 RKNN Runtime 和驱动;运行前用 rknn-smi -v 检查设备端版本。 - 源模型许可证:Apache License 2.0,正文见 LICENSE,归属与转换修改说明见 NOTICE。 - 同时遵守 RKNN Toolkit、RKNN Runtime 相关许可;本包不包含运行库。

Read RKNNAI's full model card

1. 模型介绍

本仓库提供由 Qwen/Qwen3-Reranker-0.6B 转换的 RKNN 模型。

  • Model ID:RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B
  • 模型显示名称:RK182X-EMBEDDING-Qwen3-Reranker-0.6B
  • 发布版本:v1.1.0
  • 源模型:Qwen/Qwen3-Reranker-0.6B
  • 模型类型:EMBEDDING
  • 芯片字段:RK182X
  • 具体支持芯片:RK1820、RK1828

可用模型

发布版本 配置目录 支持芯片 量化方式 NPU 核数 上下文长度(tokens) 最大文本长度(tokens) KVCache
v1.1.0 Qwen3-Reranker-0.6B-w4a16-8-2048 RK1820、RK1828 w4a16 8 2048(2k) 2048(2k) fp16

2. 文件说明

文件或目录 说明
LICENSE 源模型许可证
NOTICE 许可、使用要求及归属说明
<配置目录>/ 配套模型文件
子目录 README.md 当前配置说明
子目录 config.json 模型配置与文件清单
子目录 SHA256SUMS 当前目录交付文件的 SHA-256 校验值(不包含自身)

3. 模型下载

ModelScope 完整下载:

modelscope download --model RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B --revision v1.1.0 --local_dir ./RK182X-EMBEDDING-Qwen3-Reranker-0.6B

Hugging Face 完整下载:

hf download RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B --revision v1.1.0 --local-dir ./RK182X-EMBEDDING-Qwen3-Reranker-0.6B

ModelScope 指定配置下载:

from modelscope import snapshot_download

snapshot_download(
    "RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B",
    revision="v1.1.0",
    allow_patterns=["README.md", "LICENSE", "NOTICE", "Qwen3-Reranker-0.6B-w4a16-8-2048/**"],
    local_dir="./RK182X-EMBEDDING-Qwen3-Reranker-0.6B",
)

Hugging Face 指定配置下载:

hf download RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B --revision v1.1.0 --include "README.md" "LICENSE" "NOTICE" "Qwen3-Reranker-0.6B-w4a16-8-2048/**" --local-dir ./RK182X-EMBEDDING-Qwen3-Reranker-0.6B

4. SHA-256 校验

在配置目录执行:

cd ./RK182X-EMBEDDING-Qwen3-Reranker-0.6B/Qwen3-Reranker-0.6B-w4a16-8-2048
sha256sum -c SHA256SUMS

所有条目显示 OK 后再部署。

5. 兼容性与限制

  • 支持芯片:RK1820、RK1828。
  • 使用配套 RKNN Runtime 和驱动;运行前用 rknn-smi -v 检查设备端版本。
  • 所选配置目录内的文件须配套使用。
  • 用于文本相关性重排序,输出相关性分数。

6. 版权与许可证

  • 源模型许可证:Apache License 2.0,正文见 LICENSE,归属与转换修改说明见 NOTICE。
  • 本仓库交付物为源权重的转换衍生物,仍受源模型许可和使用限制约束;不是上游未经修改的原始权重。
  • 同时遵守 RKNN Toolkit、RKNN Runtime 相关许可;本包不包含运行库。

Identity and Version

Repository
RKNNAI/RK182X-EMBEDDING-Qwen3-Reranker-0.6B
Publisher
RKNNAI
Task
Text ranking
Modality
Other
Library
Not stated by the source
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
ffe4f37aaffe6d263f59b10bb4a594b0278d627c
First published
2026-09-24
Last updated
2026-09-28

Files and Weights

12 files, 597.8 MB in total. The weights are 2 files totalling 316.5 MB in bin, gguf.

Weights2 files · 316.5 MB
Configuration1 file · 6.5 KB
Documentation4 files · 17.4 KB
Other4 files · 281.2 MB
Repository1 file · 1.8 KB
Every file
FileTypeSizeSHA-256
Qwen3-Reranker-0.6B-w4a16-8-2048/Qwen3-Reranker-0.6B-grq.embed.binWeights310.6 MB 4db07e67017a
Qwen3-Reranker-0.6B-w4a16-8-2048/Qwen3-Reranker-0.6B-grq.tokenizer.ggufWeights5.9 MB 7616c7cd8cc2
Qwen3-Reranker-0.6B-w4a16-8-2048/config.jsonConfiguration6.5 KB —
LICENSEDocumentation11.4 KB —
NOTICEDocumentation504 B —
Qwen3-Reranker-0.6B-w4a16-8-2048/README.mdDocumentation2.3 KB —
README.mdDocumentation3.2 KB —
Qwen3-Reranker-0.6B-w4a16-8-2048/Qwen3-Reranker-0.6B-grq-model_report.htmlOther814.7 KB —
Qwen3-Reranker-0.6B-w4a16-8-2048/Qwen3-Reranker-0.6B-grq.rknnOther24.5 MB 252391e10a71
Qwen3-Reranker-0.6B-w4a16-8-2048/Qwen3-Reranker-0.6B-grq.weightOther256.0 MB d13936baa572
Qwen3-Reranker-0.6B-w4a16-8-2048/SHA256SUMSOther659 B —
.gitattributesRepository1.8 KB —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
316.5 MB
Download from RKNNAI

Released by RKNNAI through its official repository on Hugging Face. Read the license.

Built From

Memory Requirements

PrecisionWeights in memory
As published316.5 MB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About RK182X-EMBEDDING-Qwen3-Reranker-0.6B

Can I use RK182X-EMBEDDING-Qwen3-Reranker-0.6B commercially?

Yes. RK182X-EMBEDDING-Qwen3-Reranker-0.6B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

An English cross-encoder for ranking passages from notes, procedures, decisions and technical documentation. Give it a query and candidate passages; it returns one relevance score per pair. It is a fine-tune of Ettin Reranker 150M and runs locally through ONNX Runtime. This revision adds a Vulkan FP32 graph, model.vulkan.onnx; the learned weights and the CUDA FP16 graph are unchanged. It omits the DirectML graph, which remains in the previous revision. Install tokenizers, numpy and onnxruntime-gpu[cuda,cudnn]==1.24.4. Keep model.onnx, model.onnx.data and the tokenizer from the same package together. This CUDA example assumes that package is in downloaded-model and scores two passages…

Open weights apache-2.0 7,999 tokens

This model was converted to GGUF format from madebyaris/rerank-indonesia using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Install llama.cpp through brew (works on Mac and Linux) Invoke the llama.cpp server or the CLI. Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with LLAMACURL=1 flag along with other hardware-specific flags (for ex: LLAMACUDA=1 for Nvidia GPUs on Linux).

Open weights apache-2.0 sentence-transformers

Model · Text ranking

CLM-v0.1-8B-gguf

EvoAwaken Workshop

This repository contains the state and action projection heads only from CLM scores candidate actions against a state; it does not generate text. A separate, frozen Qwen3-8B encoder is required to turn text into embeddings. The source checkpoint is CLMv0.1-8B.pt (SHA-256 b2b4a8c9c2d39263eff78a351eb909a342ce9b3bf21a3f07c1d1bf15f1c4eda5). The converter preserves the F32 weight and bias bytes. It stores the reference scoring scale min(exp(logitscale), 100) = 100 in the GGUF. The GGUF does not contain the Qwen3-8B encoder, tokenizer, or original.pt file. The verified encoder file is SHA-256 5e416a2020fe63e76ea13c8979be35fc6070aaf3578f7876400c55c2f5c3eb30. Use rust-model-inference at or after…

Open weights apache-2.0

Model · Text ranking

kev-9B-LoRA-ranker

Gogogo

Why the disagreement is the feature, not the bug: kev downgrades high-frequency auto-recovering alerts (the #1 source of alert fatigue) and escalates orchestration-core outages that static rules treat as routine. - None-injection (p=0.1): teaches the head to abstain, not everything deserves a label We ran the obvious next experiment: fine-tune it on its own production data (5-fold grouped CV, 4-arm temperature calibration). Result: 0 gain over the deployed version — and we did not deploy it, per preset criteria defined before the experiment. That's the discipline this model is proud of: judge the upgrade on pre-registered criteria, not vibes. Training receipts and the arbitration table are…

Open weights apache-2.0 peft

This repository is under construction. No selected, validated model has been released yet. This model was created by Thea Tollersrud and Per Egil Kummervold in the NB-ASR project. The planned model recognizes equivalent Bokmål–Nynorsk text while detecting changed numbers, pronouns, entities, and participant roles. Meaning-preserving differences and reorderings are accepted. Training starts from Qwen/Qwen3-Reranker-0.6B and uses revision 619e36b85e4943f23cd85b66edcde331995e4872. Validation is automatic, without manual review. Accuracy will be conditional on the supplied positive pairs and the automatic corruption rules. See trainingstatus.json for the latest recorded job status. A complete…

Open weights apache-2.0

This is Horizon-Labs/multilingual-reranker-small in a smaller ONNX file, for running on a CPU. Same model, same tokenizer, same scores to within a small margin. The file is about a third smaller than the original quantized one. We use it in Sno Station to rerank memories for coding agents. The token embeddings, the attention projections and the first MLP projection are stored in 8 bits. The second MLP projection stays in float. Its inputs carry a few very large values, and 8 bits move the scores too much there. Weights are unsigned 8-bit, per channel. Signed weights overflow on AVX2 CPUs without VNNI and flatten every score. quantize.py rebuilds the file from the original onnx/model.onnx at…

Open weights apache-2.0 8,192 tokens transformers.js