Cohere Embed Multilingual v3 vs Jina Embeddings v3

Detailed comparison between Cohere Embed Multilingual v3 and Jina Embeddings v3. See which embedding best meets your accuracy and performance needs.

Model Comparison

Cohere Embed Multilingual v3 takes the lead.

Both Cohere Embed Multilingual v3 and Jina Embeddings v3 are powerful embedding models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.

Why Cohere Embed Multilingual v3:

  • Cohere Embed Multilingual v3 delivers better accuracy (nDCG@10: 0.781 vs 0.766)
  • Cohere Embed Multilingual v3 is 189739ms faster on average

Overview

Key metrics

ELO Rating

Overall ranking quality

Cohere Embed Multilingual v3

1501

Jina Embeddings v3

1491

Win Rate

Head-to-head performance

Cohere Embed Multilingual v3

42.9%

Jina Embeddings v3

40.3%

Accuracy (nDCG@10)

Ranking quality metric

Cohere Embed Multilingual v3

0.781

Jina Embeddings v3

0.766

Average Latency

Response time

Cohere Embed Multilingual v3

24024ms

Jina Embeddings v3

213763ms

Visual Performance Analysis

Performance

ELO Rating Comparison

Win/Loss/Tie Breakdown

Accuracy Across Datasets (nDCG@10)

Latency Distribution (ms)

Breakdown

How the models stack up

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Overall Performance
ELO Rating
1501
1491
Overall ranking quality based on pairwise comparisons
Win Rate
42.9%
40.3%
Percentage of comparisons won against other models
Pricing & Availability
Price per 1M tokens
$0.100
$0.045
Cost per million tokens processed
Release Date
2024-02-07
2024-09-18
Model release date
Accuracy Metrics
Avg nDCG@10
0.781
0.766
Normalized discounted cumulative gain at position 10
Performance Metrics
Avg Latency
24024ms
213763ms
Average response time across all datasets

Dataset Performance

By field

Comprehensive comparison of accuracy metrics (nDCG, Recall) and latency percentiles for each benchmark dataset.

PG

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
Latency Metrics
Mean
37785ms
155423ms
Average response time
P50
37029ms
152315ms
50th percentile (median)
P90
43453ms
178736ms
90th percentile

business reports

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
Latency Metrics
Mean
5204ms
18149ms
Average response time
P50
5100ms
17786ms
50th percentile (median)
P90
5985ms
20871ms
90th percentile

DBPedia

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
nDCG@5
0.619
0.616
Ranking quality at top 5 results
nDCG@10
0.591
0.591
Ranking quality at top 10 results
Recall@5
0.222
0.225
% of relevant docs in top 5
Recall@10
0.329
0.340
% of relevant docs in top 10
Latency Metrics
Mean
33570ms
199515ms
Average response time
P50
32899ms
195525ms
50th percentile (median)
P90
38606ms
229442ms
90th percentile

FiQa

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
nDCG@5
0.637
0.645
Ranking quality at top 5 results
nDCG@10
0.654
0.679
Ranking quality at top 10 results
Recall@5
0.621
0.633
% of relevant docs in top 5
Recall@10
0.692
0.741
% of relevant docs in top 10
Latency Metrics
Mean
37581ms
458812ms
Average response time
P50
36829ms
449636ms
50th percentile (median)
P90
43218ms
527634ms
90th percentile

SciFact

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
nDCG@5
0.723
0.622
Ranking quality at top 5 results
nDCG@10
0.732
0.661
Ranking quality at top 10 results
Recall@5
0.808
0.695
% of relevant docs in top 5
Recall@10
0.833
0.800
% of relevant docs in top 10
Latency Metrics
Mean
39542ms
270629ms
Average response time
P50
38751ms
265216ms
50th percentile (median)
P90
45473ms
311223ms
90th percentile

MSMARCO

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
nDCG@5
0.996
1.000
Ranking quality at top 5 results
nDCG@10
0.994
0.997
Ranking quality at top 10 results
Recall@5
0.122
0.123
% of relevant docs in top 5
Recall@10
0.218
0.220
% of relevant docs in top 10
Latency Metrics
Mean
32380ms
476020ms
Average response time
P50
31732ms
466500ms
50th percentile (median)
P90
37237ms
547423ms
90th percentile

NorQuAD

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
Latency Metrics
Mean
3824ms
45614ms
Average response time
P50
3748ms
44702ms
50th percentile (median)
P90
4398ms
52456ms
90th percentile

ARCD

MetricCohere Embed Multilingual v3Jina Embeddings v3Description
Accuracy Metrics
nDCG@5
0.925
0.888
Ranking quality at top 5 results
nDCG@10
0.933
0.900
Ranking quality at top 10 results
Recall@5
0.940
0.920
% of relevant docs in top 5
Recall@10
0.960
0.960
% of relevant docs in top 10
Latency Metrics
Mean
2305ms
85941ms
Average response time
P50
2259ms
84222ms
50th percentile (median)
P90
2651ms
98832ms
90th percentile

Explore More

Compare more embeddings

See how all embedding models stack up. Compare OpenAI, Cohere, Jina AI, Voyage, and more. View comprehensive benchmarks, compare performance metrics, and find the perfect embedding for your RAG application.