ONNX Model Zoo

llmware/jina-reranker-tiny-onnx

By llmware · ↓ 4 · ♥ 0 · Updated 2025-02-11

What's in this repo

This page lists the files in llmware/jina-reranker-tiny-onnx, including model.onnx, model_bnb4.onnx, model_fp16.onnx and more.

  • 7 ONNX models (.onnx files)
  • 14 files, including 7 .onnx, 5 .json, 1 .md
  • Last updated 2025-02-11

ONNX files

File Size Download
model.onnx — Download
model_bnb4.onnx — Download
model_fp16.onnx — Download
model_int8.onnx — Download
model_q4.onnx — Download
model_quantized.onnx — Download
model_uint8.onnx — Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

jina-reranker-tiny-onnx

jina-reranker-tiny-onnx is an ONNX packaging of Jinaai/jina-reranker-v1-tiny-en, providing a very small, very fast inference implementation, optimized for AI PCs.

jina-reranker-tiny-onnx is a tiny specialized semantic reranker that ranks the similarity between a query and a text passage.

Model Description

  • Developed by: jinaai
  • Quantized by: llmware
  • Model type: jina-reranker-v1-tiny-en
  • Parameters: 65 million
  • Model Parent: jinaai/jina-reranker-v1-tiny-en
  • Language(s) (NLP): English
  • License: Apache 2.0
  • Uses: Semantic similarity classification and ranking
  • RAG Benchmark Accuracy Score: NA
  • Quantization: int4

Model Card Contact

llmware on github

llmware on hf

llmware website

Model info

onnx bert green p1 llmware-semantic onns custom_code base_model:jinaai/jina-reranker-v1-tiny-en base_model:quantized:jinaai/jina-reranker-v1-tiny-en license:apache-2.0 region:us
↓ 4 ♥ 0

Created: 2025-01-12

Updated: 2025-02-11

View on Hugging Face

Files (14)

.gitattributes
README.md
config.json
model.onnx ONNX
model_bnb4.onnx ONNX
model_fp16.onnx ONNX
model_int8.onnx ONNX
model_q4.onnx ONNX
model_quantized.onnx ONNX
model_uint8.onnx ONNX
special_tokens_map.json
tokenizer.json
tokenizer_config.json
vocab.json

Related models