multilingual-e5-large-instruct-Q5_K_M-GGUF: An Open-Source Multilingual Instruction Model - Supports Classification, Retrieval, and Clustering Tasks

Multilingual E5 Large Instruct Q5 K M GGUF

Developed by JHJHJHJHJ

Multilingual E5 large instruction model, supporting multiple languages and tasks, including classification, retrieval, clustering, etc.

Large Language Model Supports Multiple LanguagesOpen Source License:MIT #Multilingual Text Embedding #High-Precision Semantic Retrieval #Cross-Language Classification

Downloads 18

Release Time : 1/21/2025

Model Overview

This model is a multilingual embedding model based on intfloat/multilingual-e5-large-instruct, supporting a wide range of natural language processing tasks, including text classification, information retrieval, and text clustering.

Model Features

Multilingual Support

Supports over 100 languages, suitable for global multilingual application scenarios.

High Performance

Outstanding performance in multiple benchmarks, especially in text classification and information retrieval tasks.

Versatile

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

Model Capabilities

Text classification

Information retrieval

Text clustering

Bilingual mining

Sentence embedding

Multilingual processing

Use Cases

E-commerce

Product Review Classification

Performs sentiment classification on Amazon product reviews to distinguish between positive and negative feedback.

Achieves an accuracy of 96.29% in the MTEB AmazonPolarityClassification task.

Multilingual Review Analysis

Supports classification of Amazon product reviews in multiple languages, including English, German, Spanish, etc.

Accuracy rates: 56.72% for English, 52.99% for German, and 48.81% for Spanish.

Information Retrieval

Document Retrieval

Retrieves relevant information from large-scale document libraries.

Achieves a map@100 of 49.88% in the MTEB ArguAna task.

Question-Answering Systems

Supports information retrieval functionality in QA systems.

Achieves an mrr@10 of 82.17% in the MTEB HotpotQA task.

Bilingual Applications

Bilingual Text Mining

Identifies corresponding text segments in different languages.

Achieves an accuracy of 99.66% for German-English pairs in the MTEB BUCC task.

🚀 multilingual-e5-large-instruct

This is a multilingual model based on intfloat/multilingual-e5-large-instruct, with performance metrics across multiple tasks and datasets.

📚 Documentation

Model Information

Property	Details
Model Type	multilingual-e5-large-instruct
Base Model	intfloat/multilingual-e5-large-instruct
Tags	mteb, sentence-transformers, transformers, llama-cpp, gguf-my-repo
Supported Languages	multilingual, af, am, ar, as, az, be, bg, bn, br, bs, ca, cs, cy, da, de, el, en, eo, es, et, eu, fa, fi, fr, fy, ga, gd, gl, gu, ha, he, hi, hr, hu, hy, id, is, it, ja, jv, ka, kk, km, kn, ko, ku, ky, la, lo, lt, lv, mg, mk, ml, mn, mr, ms, my, ne, nl, 'no', om, or, pa, pl, ps, pt, ro, ru, sa, sd, si, sk, sl, so, sq, sr, su, sv, sw, ta, te, th, tl, tr, ug, uk, ur, uz, vi, xh, yi, zh
License	mit

Model Performance Results

The model has been tested on various tasks and datasets, with the following performance metrics:

Classification Tasks

Dataset Name	Accuracy	AP	F1
MTEB AmazonCounterfactualClassification (en)	76.23880597014924	39.07351965022687	70.04836733862683
MTEB AmazonCounterfactualClassification (de)	66.71306209850107	79.01499914759529	64.81951817560703
MTEB AmazonCounterfactualClassification (en-ext)	73.85307346326837	22.447519885878737	61.0162730745633
MTEB AmazonCounterfactualClassification (ja)	76.04925053533191	23.44983217128922	62.5723230907759
MTEB AmazonPolarityClassification	96.28742500000001	94.8449918887462	96.28680923610432
MTEB AmazonReviewsClassification (en)	56.716	-	55.76510398266401
MTEB AmazonReviewsClassification (de)	52.99999999999999	-	52.00829994765178
MTEB AmazonReviewsClassification (es)	48.806000000000004	-	48.082345914983634
MTEB AmazonReviewsClassification (fr)	48.507999999999996	-	47.68752844642045
MTEB AmazonReviewsClassification (ja)	47.709999999999994	-	47.05870376637181
MTEB AmazonReviewsClassification (zh)	44.662000000000006	-	43.42371965372771
MTEB Banking77Classification	85.73376623376623	-	85.68480707214599
MTEB EmotionClassification	51.51	-	47.632159862049896

Retrieval Tasks

Dataset Name	MAP@1	MAP@10	MAP@100	MAP@1000	MAP@3	MAP@5	MRR@1	MRR@10	MRR@100	MRR@1000	MRR@3	MRR@5	NDCG@1	NDCG@10	NDCG@100	NDCG@1000	NDCG@3	NDCG@5	Precision@1	Precision@10	Precision@100	Precision@1000	Precision@3	Precision@5	Recall@1	Recall@10	Recall@100	Recall@1000	Recall@3	Recall@5
MTEB ArguAna	31.721	49.221	49.884	49.888	44.31	47.276	32.432	49.5	50.163000000000004	50.166	44.618	47.541	31.721	58.384	61.111000000000004	61.187999999999995	48.386	53.708999999999996	31.721	8.741	0.991	0.1	20.057	14.609	31.721	87.411	99.075	99.644	60.171	73.044
MTEB CQADupstackRetrieval	27.764166666666668	37.298166666666674	38.530166666666666	38.64416666666667	34.484833333333334	36.0385	32.93558333333333	41.589749999999995	42.425333333333334	42.476333333333336	39.26825	40.567083333333336	32.93558333333333	42.706583333333334	47.82483333333333	49.95733333333334	38.064750000000004	40.18158333333333	32.93558333333333	7.459833333333334	1.1830833333333335	0.15608333333333332	17.5235	12.349833333333333	27.764166666666668	54.31775	76.74350000000001	91.45208333333332	41.23425	46.73983333333334
MTEB ClimateFEVER	12.969	21.584999999999997	23.3	23.5	18.218999999999998	19.983	29.316	40.033	40.96	41.001	37.123	38.757999999999996	29.316	29.858	36.756	40.245999999999995	24.822	26.565	29.316	9.186	1.6549999999999998	0.22999999999999998	18.436	13.876	12.969	35.142	59.143	78.594	22.604	27.883000000000003
MTEB DBPedia	8.527999999999999	17.974999999999998	25.665	27.406000000000002	13.017999999999999	15.137	62.5	71.891	72.294	72.296	69.958	71.121	50.875	38.36	44.235	52.154	43.008	40.083999999999996	62.5	30.0	10.038	2.0869999999999997	46.833000000000006	38.800000000000004	8.527999999999999	23.828	52.322	77.143	14.136000000000001	17.761
MTEB FEVER	60.734	72.442	72.735	72.75	70.41199999999999	71.80499999999999	65.212	76.613	76.79899999999999	76.801	74.8	76.12400000000001	65.212	77.988	79.167	79.452	74.362	76.666	65.212	10.003	1.077	0.11199999999999999	29.518	19.016	60.734	90.824	95.71600000000001	97.577	81.243	86.90299999999999
MTEB FiQA2018	23.845	39.281	41.422	41.593	34.467	37.017	47.531	56.204	56.928999999999995	56.962999999999994	54.115	55.373000000000005	47.531	47.711999999999996	54.510999999999996	57.103	44.145	45.032	47.531	13.194	2.045	0.249	29.424	21.451	23.845	54.967	79.11399999999999	94.56700000000001	40.256	46.215
MTEB HotpotQA	37.819	60.889	61.717999999999996	61.778	57.254000000000005	59.541	75.638	82.173	82.362	82.37	81.089	81.827	75.638	69.317	72.221	73.382	64.14	67.07600000000001	75.638	14.704999999999998	1.698	0.185	41.394999999999996	27.162999999999997	37.819	60.889	61.717999999999996	61.778	57.254000000000005	59.541

Clustering Tasks

Dataset Name	V-Measure
MTEB ArxivClusteringP2P	46.40419580759799
MTEB ArxivClusteringS2S	40.48593255007969
MTEB BiorxivClusteringP2P	40.935218072113855
MTEB BiorxivClusteringS2S	36.276389017675264

Reranking Tasks

Dataset Name	MAP	MRR
MTEB AskUbuntuDupQuestions	63.889179122289995	77.61146286769556

STS Tasks

Dataset Name	Cosine Similarity (Pearson)	Cosine Similarity (Spearman)	Euclidean Distance (Pearson)	Euclidean Distance (Spearman)	Manhattan Distance (Pearson)	Manhattan Distance (Spearman)
MTEB BIOSSES	88.15075203727929	86.9622224570873	86.70473853624121	86.9622224570873	86.21089380980065	86.75318154937008

Bitext Mining Tasks

Dataset Name	Accuracy	F1	Precision	Recall
MTEB BUCC (de-en)	99.65553235908142	99.60681976339595	99.58246346555325	99.65553235908142
MTEB BUCC (fr-en)	99.26260180497468	99.14520507740848	99.08650671362535	99.26260180497468
MTEB BUCC (ru-en)	98.07412538967787	97.86629719431936	97.76238309664012	98.07412538967787
MTEB BUCC (zh-en)	99.42074776197998	99.38564156573635	99.36808846761454	99.42074776197998

📄 License

This project is licensed under the MIT 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