Granite-Embedding-107M-Multilingual-Onnx Open-Source Multilingual Embedding Model

Granite Embedding 107m Multilingual Onnx

Developed by gety-ai

A 107M-parameter multilingual embedding model developed by IBM, supporting 12 languages with excellent performance on the MTEB benchmark

Text Embedding

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Multilingual Embedding #Text Classification #Low-resource Optimization

Downloads 53

Release Time : 2/19/2025

Model Overview

This model is a multilingual text embedding model capable of converting texts in different languages into high-quality vector representations, suitable for tasks like cross-lingual retrieval and classification

Model Features

Multilingual Support

Supports text embedding for 12 major languages, enabling cross-lingual semantic understanding

Efficient Performance

Lightweight design with 107M parameters, improving inference efficiency while maintaining high quality

MTEB Benchmark Validation

Demonstrates excellent performance across multiple MTEB benchmark tasks, validating model reliability

Model Capabilities

Text Vectorization

Cross-lingual Semantic Retrieval

Text Classification

Semantic Similarity Calculation

Use Cases

E-commerce

Multilingual Product Review Analysis

Analyze product reviews from users in different languages

Achieved 36.41% accuracy on Amazon review classification task

Sentiment Analysis

Identify sentiment tendencies in user reviews

Achieved 66.58% accuracy on Amazon sentiment classification task

Information Retrieval

Cross-lingual Document Retrieval

Retrieve relevant documents in other languages using queries in one language

Achieved NDCG@10 of 4.537 on AppsRetrieval task

🚀 ibm-granite/granite-embedding-107m-multilingual

This is a multilingual embedding model from transformers library. It supports multiple languages and has been tested on various MTEB datasets, showing performance in tasks like classification, retrieval, and clustering.

📚 Documentation

Supported Languages

The model supports the following languages:

English (en)
Arabic (ar)
Czech (cs)
German (de)
Spanish (es)
French (fr)
Italian (it)
Japanese (ja)
Korean (ko)
Dutch (nl)
Portuguese (pt)
Chinese (zh)

Model Performance

1. MTEB AmazonCounterfactualClassification

Dataset Config	Task Type	Accuracy	F1	F1 Weighted	AP	AP Weighted	Main Score
en-ext	Classification	72.7136	60.44540000000001	77.8541	22.4958	22.4958	72.7136
en	Classification	71.6716	65.4221	74.3533	33.7567	33.7567	71.6716

2. MTEB AmazonPolarityClassification

Dataset Config	Task Type	Accuracy	F1	F1 Weighted	AP	AP Weighted	Main Score
default	Classification	66.5804	66.2191	66.2191	61.340799999999994	61.340799999999994	66.5804

3. MTEB AmazonReviewsClassification

Dataset Config	Task Type	Accuracy	F1	F1 Weighted	Main Score
en	Classification	36.412	35.633199999999995	35.633199999999995	36.412

4. MTEB AppsRetrieval

| Dataset Config | Task Type | NDCG@1 | NDCG@3 | NDCG@5 | NDCG@10 | NDCG@20 | NDCG@100 | NDCG@1000 | MAP@1 | MAP@3 | MAP@5 | MAP@10 | MAP@20 | MAP@100 | MAP@1000 | Recall@1 | Recall@3 | Recall@5 | Recall@10 | Recall@20 | Recall@100 | Recall@1000 | Precision@1 | Precision@3 | Precision@5 | Precision@10 | Precision@20 | Precision@100 | Precision@1000 | MRR@1 | MRR@3 | MRR@5 | MRR@10 | MRR@20 | MRR@100 | MRR@1000 | NAUC NDCG@1 Max | NAUC NDCG@1 Std | NAUC NDCG@1 Diff1 | NAUC NDCG@3 Max | NAUC NDCG@3 Std | NAUC NDCG@3 Diff1 | NAUC NDCG@5 Max | NAUC NDCG@5 Std | NAUC NDCG@5 Diff1 | NAUC NDCG@10 Max | NAUC NDCG@10 Std | NAUC NDCG@10 Diff1 | NAUC NDCG@20 Max | NAUC NDCG@20 Std | NAUC NDCG@20 Diff1 | NAUC NDCG@100 Max | NAUC NDCG@100 Std | NAUC NDCG@100 Diff1 | NAUC NDCG@1000 Max | NAUC NDCG@1000 Std | NAUC NDCG@1000 Diff1 | NAUC MAP@1 Max | NAUC MAP@1 Std | NAUC MAP@1 Diff1 | NAUC MAP@3 Max | NAUC MAP@3 Std | NAUC MAP@3 Diff1 | NAUC MAP@5 Max | NAUC MAP@5 Std | NAUC MAP@5 Diff1 | NAUC MAP@10 Max | NAUC MAP@10 Std | NAUC MAP@10 Diff1 | NAUC MAP@20 Max | NAUC MAP@20 Std | NAUC MAP@20 Diff1 | NAUC MAP@100 Max | NAUC MAP@100 Std | NAUC MAP@100 Diff1 | NAUC MAP@1000 Max | NAUC MAP@1000 Std | NAUC MAP@1000 Diff1 | NAUC Recall@1 Max | NAUC Recall@1 Std | NAUC Recall@1 Diff1 | NAUC Recall@3 Max | NAUC Recall@3 Std | NAUC Recall@3 Diff1 | NAUC Recall@5 Max | NAUC Recall@5 Std | NAUC Recall@5 Diff1 | NAUC Recall@10 Max | NAUC Recall@10 Std | NAUC Recall@10 Diff1 | NAUC Recall@20 Max | NAUC Recall@20 Std | NAUC Recall@20 Diff1 | NAUC Recall@100 Max | NAUC Recall@100 Std | NAUC Recall@100 Diff1 | NAUC Recall@1000 Max | NAUC Recall@1000 Std | NAUC Recall@1000 Diff1 | NAUC Precision@1 Max | NAUC Precision@1 Std | NAUC Precision@1 Diff1 | NAUC Precision@3 Max | NAUC Precision@3 Std | NAUC Precision@3 Diff1 | NAUC Precision@5 Max | NAUC Precision@5 Std | NAUC Precision@5 Diff1 | NAUC Precision@10 Max | NAUC Precision@10 Std | NAUC Precision@10 Diff1 | NAUC Precision@20 Max | NAUC Precision@20 Std | NAUC Precision@20 Diff1 | NAUC Precision@100 Max | NAUC Precision@100 Std | NAUC Precision@100 Diff1 | NAUC Precision@1000 Max | NAUC Precision@1000 Std | NAUC Precision@1000 Diff1 | NAUC MRR@1 Max | NAUC MRR@1 Std | NAUC MRR@1 Diff1 | NAUC MRR@3 Max | NAUC MRR@3 Std | NAUC MRR@3 Diff1 | NAUC MRR@5 Max | NAUC MRR@5 Std | NAUC MRR@5 Diff1 | NAUC MRR@10 Max | NAUC MRR@10 Std | NAUC MRR@10 Diff1 | NAUC MRR@20 Max | NAUC MRR@20 Std | NAUC MRR@20 Diff1 | NAUC MRR@100 Max | NAUC MRR@100 Std | NAUC MRR@100 Diff1 | NAUC MRR@1000 Max | NAUC MRR@1000 Std | NAUC MRR@1000 Diff1 | Main Score | |----------------|-----------|--------|--------|--------|---------|---------|----------|-----------|-------|-------|-------|--------|--------|---------|----------|---------|---------|---------|----------|----------|-----------|-----------|------------|------------|-----------|-----------|-----------|------------|------------|-------------|-------------|-------|-------|-------|--------|--------|---------|----------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|------------| | default | Retrieval | 2.39 | 3.527 | 3.9759999999999995 | 4.537 | 5.140000000000001 | 6.526 | 9.797 | 2.39 | 3.2489999999999997 | 3.499 | 3.7220000000000004 | 3.887 | 4.058 | 4.146 | 2.39 | 4.329000000000001 | 5.418 | 7.198 | 9.588000000000001 | 17.371 | 45.206 | 2.39 | 1.443 | 1.084 | 0.72 | 0.479 | 0.174 | 0.045 | 2.3904 | 3.2492 | 3.4989 | 3.7220000000000004 | 3.8869000000000002 | 4.0578 | 4.1463 | 37.599700000000006 | 20.302899999999998 | 40.4987 | 31.119400000000002 | 11.7335 | 28.788000000000004 | 28.505399999999998 | 12.1402 | 25.730900000000002 | 27.0656 | 12.648699999999998 | 22.0832 | 25.953599999999998 | 12.550500000000001 | 19.3722 | 23.268 | 12.8176 | 15.9275 | 21.921499999999998 | 12.656300000000002 | 13.9004 | 37.599700000000006 | 20.302899999999998 | 40.4987 | 32.2818 | 13.276399999999999 | 30.9064 | 30.5166 | 13.406 | 28.8213 | 29.731999999999996 | 13.5688 | 26.888499999999997 | 29.211399999999998 | 13.4739 | 25.6814 | 28.578300000000002 | 13.5385 | 24.793100000000003 | 28.3912 | 13.5039 | 24.570600000000002 | 37.599700000000006 | 20.302899999999998 | 40.4987 | 28.598000000000003 | 8.3847 | 24.1871 | 24.5381 | 9.8274 | 19.6821 | 22.5445 | 11.4415 | 13.8268 | 21.3196 | 11.5932 | 10.1991 | 16.9415 | 12.353200000000001 | 5.7534 | 15.9223 | 12.2848 | 3.5477000000000003 | 37.599700000000006 | 20.302899999999998 | 40.4987 | 28.598000000000003 | 8.3847 | 24.1871 | 24.5381 | 9.8274 | 19.6821 | 22.5445 | 11.4415 | 13.8268 | 21.3196 | 11.5932 | 10.1991 | 16.9415 | 12.353200000000001 | 5.7534 | 15.9223 | 12.2848 | 3.5477000000000003 | 37.599700000000006 | 20.302899999999998 | 40.4987 | 32.2818 | 13.276399999999999 | 30.9064 | 30.5166 | 13.406 | 28.8213 | 29.731999999999996 | 13.5688 | 26.888499999999997 | 29.211399999999998 | 13.4739 | 25.6814 | 28.578300000000002 | 13.5385 | 24.793100000000003 | 28.3912 | 13.5039 | 24.570600000000002 | 4.537 |

5. MTEB ArguAna

| Dataset Config | Task Type | NDCG@1 | NDCG@3 | NDCG@5 | NDCG@10 | NDCG@20 | NDCG@100 | NDCG@1000 | MAP@1 | MAP@3 | MAP@5 | MAP@10 | MAP@20 | MAP@100 | MAP@1000 | Recall@1 | Recall@3 | Recall@5 | Recall@10 | Recall@20 | Recall@100 | Recall@1000 | Precision@1 | Precision@3 | Precision@5 | Precision@10 | Precision@20 | Precision@100 | Precision@1000 | MRR@1 | MRR@3 | MRR@5 | MRR@10 | MRR@20 | MRR@100 | MRR@1000 | NAUC NDCG@1 Max | NAUC NDCG@1 Std | NAUC NDCG@1 Diff1 | NAUC NDCG@3 Max | NAUC NDCG@3 Std | NAUC NDCG@3 Diff1 | NAUC NDCG@5 Max | NAUC NDCG@5 Std | NAUC NDCG@5 Diff1 | NAUC NDCG@10 Max | NAUC NDCG@10 Std | NAUC NDCG@10 Diff1 | NAUC NDCG@20 Max | NAUC NDCG@20 Std | NAUC NDCG@20 Diff1 | NAUC NDCG@100 Max | NAUC NDCG@100 Std | NAUC NDCG@100 Diff1 | NAUC NDCG@1000 Max | NAUC NDCG@1000 Std | NAUC NDCG@1000 Diff1 | NAUC MAP@1 Max | NAUC MAP@1 Std | NAUC MAP@1 Diff1 | NAUC MAP@3 Max | NAUC MAP@3 Std | NAUC MAP@3 Diff1 | NAUC MAP@5 Max | NAUC MAP@5 Std | NAUC MAP@5 Diff1 | NAUC MAP@10 Max | NAUC MAP@10 Std | NAUC MAP@10 Diff1 | NAUC MAP@20 Max | NAUC MAP@20 Std | NAUC MAP@20 Diff1 | NAUC MAP@100 Max | NAUC MAP@100 Std | NAUC MAP@100 Diff1 | NAUC MAP@1000 Max | NAUC MAP@1000 Std | NAUC MAP@1000 Diff1 | NAUC Recall@1 Max | NAUC Recall@1 Std | NAUC Recall@1 Diff1 | NAUC Recall@3 Max | NAUC Recall@3 Std | NAUC Recall@3 Diff1 | NAUC Recall@5 Max | NAUC Recall@5 Std | NAUC Recall@5 Diff1 | NAUC Recall@10 Max | NAUC Recall@10 Std | NAUC Recall@10 Diff1 | NAUC Recall@20 Max | NAUC Recall@20 Std | NAUC Recall@20 Diff1 | NAUC Recall@100 Max | NAUC Recall@100 Std | NAUC Recall@100 Diff1 | NAUC Recall@1000 Max | NAUC Recall@1000 Std | NAUC Recall@1000 Diff1 | NAUC Precision@1 Max | NAUC Precision@1 Std | NAUC Precision@1 Diff1 | NAUC Precision@3 Max | NAUC Precision@3 Std | NAUC Precision@3 Diff1 | NAUC Precision@5 Max | NAUC Precision@5 Std | NAUC Precision@5 Diff1 | NAUC Precision@10 Max | NAUC Precision@10 Std | NAUC Precision@10 Diff1 | NAUC Precision@20 Max | NAUC Precision@20 Std | NAUC Precision@20 Diff1 | NAUC Precision@100 Max | NAUC Precision@100 Std | NAUC Precision@100 Diff1 | NAUC Precision@1000 Max | NAUC Precision@1000 Std | NAUC Precision@1000 Diff1 | NAUC MRR@1 Max | NAUC MRR@1 Std | NAUC MRR@1 Diff1 | NAUC MRR@3 Max | NAUC MRR@3 Std | NAUC MRR@3 Diff1 | NAUC MRR@5 Max | NAUC MRR@5 Std | NAUC MRR@5 Diff1 | NAUC MRR@10 Max | NAUC MRR@10 Std | NAUC MRR@10 Diff1 | NAUC MRR@20 Max | NAUC MRR@20 Std | NAUC MRR@20 Diff1 | NAUC MRR@100 Max | NAUC MRR@100 Std | NAUC MRR@100 Diff1 | NAUC MRR@1000 Max | NAUC MRR@1000 Std | NAUC MRR@1000 Diff1 | Main Score | |----------------|-----------|--------|--------|--------|---------|---------|----------|-----------|-------|-------|-------|--------|--------|---------|----------|---------|---------|---------|----------|----------|-----------|-----------|------------|------------|-----------|-----------|-----------|------------|------------|-------------|-------------|-------|-------|-------|--------|--------|---------|----------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|-----------------|-----------------|------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|------------------|------------------|-------------------|-------------------|-------------------|------------| | default | Retrieval | 27.168999999999997 | 41.08 | 46.375 | 51.663000000000004 | 54.339999999999996 | 55.656000000000006 | 55.875 | 27.168999999999997 | 37.482 | 40.416000000000004 | 42.624 | 43.376999999999995 | 43.578 | 43.588 | 27.168999999999997 | 51.565000000000005 | 64.43799999999999 | 80.654 | 91.11 | 97.937 | 99.57300000000001 | 27.168999999999997 | 17.188 | 12.888 | 8.065 | 4.555 | 0.979 | 0.1 | 27.6671 | 37.6245 | 40.6188 | 42.8016 | 43.5582 | 43.7551 | 43.765 | -4.3233 | -3.5458000000000003 | 10.8118 | -1.1566 | -2.5897 | 8.3298 | -1.399 | -1.9604 | 7.6803 | 0.7746000000000001 | -0.9521 | 9.1107 | 1.0111999999999999 | 0.1519 | 9.5802 | -0.3616 | -0.6704 | 9.2401 | -0.6766 | -1.0513 | 9.0561 | -4.3233 | -3.5458000000000003 | 10.8118 | -1.9845000000000002 | -2.6683 | 8.7329 | -2.1342 | -2.3612 | 8.4139 | -1.331 | -1.982 | 9.004199999999999 | -1.3376000000000001 | -1.7424 | 9.1012 | -1.5152 | -1.8418 | 9.0513 | -1.5264 | -1.8530000000000002 | 9.043800000000001 | -4.3233 | -3.5458000000000003 | 10.8118 | 1.2361 | -2.4248 | 7.2543 | 0.9835999999999999 | -0.5726 | 5.2376 | 12.7099 | 4.9688 | 10.5016 | 28.2615 | 23.7662 | 17.6392 | 31.295099999999998 | 47.1556 | 24.055699999999998 | 14.418000000000001 | 56.899699999999996 | 3.7199999999999998 | -4.3233 | -3.5458000000000003 | 10.8118 | 1.2361 | -2.4248 | 7.2543 | 0.9835999999999999 | -0.5726 | 5.2376 | 12.7099 | 4.9688 | 10.5016 | 28.2615 | 23.7662 | 17.6392 | 31.295099999999998 | 47.1556 | 24.055699999999998 | 14.418000000000001 | 56.899699999999996 | 3.7199999999999998 | -4.2285 | -2.9951 | 9.2226 | -2.8361 | -2.5372 | 7.205 | -2.827 | -2.1469 | 6.9564 | -2.0531 | -1.8227 | 7.500500000000001 | -2.0823 | -1.585 | 7.5577000000000005 | -2.2609 | -1.6787 | 7.500500000000001 | -2.2721999999999998 | -1.6898 | 7.492400000000001 | 51.663000000000004 |

6. MTEB ArxivClusteringP2P

Dataset Config	Task Type	V-Measure	V-Measure Std	Main Score
default	Clustering	41.4944	13.6458	41.4944

7. MTEB ArxivClusteringS2S

Dataset Config	Task Type	V-Measure	V-Measure Std	Main Score
default	Clustering	30.6155	14.377999999999998	30.6155

8. MTEB AskUbuntuDupQuestions

Dataset Config	Task Type	MAP	MRR
default	Reranking	61.9001	[MRR value not fully provided in the original]

📄 License

This model is released 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