Jina-embeddings-GGUF Open-Source English Sentence Embedding Model - Efficiently Handle Similarity and Feature Extraction

Jina Embeddings GGUF

Developed by narainp

Jina Embeddings V2 Base is an efficient English sentence embedding model, focusing on sentence similarity and feature extraction tasks.

Text Embedding EnglishOpen Source License:Apache-2.0 #English sentence embeddings #High-precision text similarity #Multi-task evaluation benchmark

Downloads 139

Release Time : 1/8/2025

Model Overview

This model, based on Jina AI's technology, is designed to generate high-quality sentence embeddings suitable for tasks such as text similarity calculation, information retrieval, and clustering analysis.

Model Features

Efficient Sentence Embeddings

Capable of quickly generating high-quality sentence embeddings, suitable for large-scale text processing.

Multi-task Support

Supports various natural language processing tasks, including classification, clustering, and retrieval.

High Performance

Demonstrates outstanding performance in multiple benchmark tests, especially in text similarity tasks.

Model Capabilities

Sentence embedding generation

Text similarity calculation

Information retrieval

Text classification

Text clustering

Use Cases

Information Retrieval

Document Retrieval

Use sentence embeddings to quickly retrieve relevant documents.

Performs excellently on the MTEB ArguAna dataset

Text Classification

Sentiment Analysis

Classify text by sentiment polarity.

Achieves 88.54% accuracy on MTEB AmazonPolarityClassification

Text Clustering

Academic Paper Clustering

Cluster academic papers by topic.

Achieves a V-measure of 45.39 on MTEB ArxivClusteringP2P

🚀 jina-embedding-b-en-v2

This is a model for sentence embeddings, which can be used for tasks such as classification, retrieval, clustering, reranking, and semantic text similarity. It is based on the base model jinaai/jina-embeddings-v2-base-en and trained on the allenai/c4 dataset.

✨ Features

Multi - task Performance: Demonstrates performance across multiple natural language processing tasks, including Classification, Retrieval, Clustering, Reranking, and STS.
High - Quality Embeddings: Trained on large - scale datasets to generate high - quality sentence embeddings.

📚 Documentation

Model Information

Property	Details
Model Type	jina-embedding-b-en-v2
Base Model	jinaai/jina-embeddings-v2-base-en
Training Data	allenai/c4
License	apache-2.0

Evaluation Results

The model has been evaluated on various datasets, and here are some of the key results:

Classification Tasks

MTEB AmazonCounterfactualClassification (en):
- Accuracy: 74.73134328358209
- AP: 37.765427081831035
- F1: 68.79367444339518
MTEB AmazonPolarityClassification:
- Accuracy: 88.544275
- AP: 84.61328675662887
- F1: 88.51879035862375
MTEB AmazonReviewsClassification (en):
- Accuracy: 45.263999999999996
- F1: 43.778759656699435
MTEB Banking77Classification:
- Accuracy: 84.01298701298701
- F1: 83.31690714969382

Retrieval Tasks

MTEB ArguAna: Multiple metrics are reported, such as MAP@1, MAP@10, MRR@1, etc. For example, MAP@1 is 21.693, and MRR@1 is 21.977.
MTEB CQADupstackAndroidRetrieval: Various retrieval - related metrics are provided, e.g., MAP@1 is 31.391999999999996, and MRR@1 is 37.196.

Clustering Tasks

MTEB ArxivClusteringP2P: V - measure is 45.39242428696777.
MTEB ArxivClusteringS2S: V - measure is 36.675626784714.

Reranking Tasks

MTEB AskUbuntuDupQuestions: MAP is 62.247725694904034, and MRR is 74.91359978894604.

STS Tasks

MTEB BIOSSES: Multiple similarity - related metrics are reported, such as cos_sim_pearson (82.68003802970496) and cos_sim_spearman (81.23438110096286).

📄 License

This project is licensed under the apache-2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご