Agentset

1.9k

Gemini Embedding 2 vs Cohere Embed Multilingual v3

Detailed comparison between Gemini Embedding 2 and Cohere Embed Multilingual v3. See which embedding best meets your accuracy and performance needs. If you want to compare these models on your data, try Agentset.

Explore Gemini Embedding 2Explore Cohere Embed Multilingual v3

Model Comparison

Gemini Embedding 2 takes the lead.

Both Gemini Embedding 2 and Cohere Embed Multilingual v3 are powerful embedding models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.

Why Gemini Embedding 2:

  • Gemini Embedding 2 has 93 higher ELO rating
  • Cohere Embed Multilingual v3 delivers better accuracy (nDCG@10: 0.701 vs 0.628)
  • Cohere Embed Multilingual v3 is 427ms faster on average
  • Gemini Embedding 2 has a 11.1% higher win rate

Overview

Key metrics

ELO Rating

Overall ranking quality

Gemini Embedding 2

1605

Cohere Embed Multilingual v3

1512

Win Rate

Head-to-head performance

Gemini Embedding 2

59.5%

Cohere Embed Multilingual v3

48.4%

Accuracy (nDCG@10)

Ranking quality metric

Gemini Embedding 2

0.628

Cohere Embed Multilingual v3

0.701

Average Latency

Response time

Gemini Embedding 2

435ms

Cohere Embed Multilingual v3

7ms

Embedding Models Are Just One Piece of RAG

Agentset gives you a managed RAG pipeline with the top-ranked models and best practices baked in. No infrastructure to maintain, no embeddings to manage.

Get started freeSchedule demo

Trusted by teams building production RAG applications

5M+
Documents
1,500+
Teams
99.9%
Uptime

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

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Overall Performance
ELO Rating
1605
1512
Overall ranking quality based on pairwise comparisons
Win Rate
59.5%
48.4%
Percentage of comparisons won against other models
Pricing & Availability
Price per 1M tokens
$0.000
$0.100
Cost per million tokens processed
Dimensions
3072
512
Vector embedding dimensions (lower is more efficient)
Release Date
2026-03-10
2024-02-07
Model release date
Accuracy Metrics
Avg nDCG@10
0.628
0.701
Normalized discounted cumulative gain at position 10
Performance Metrics
Avg Latency
435ms
7ms
Average response time across all datasets

Build RAG in Minutes, Not Months

Agentset gives you a complete RAG API with top-ranked embedding models and smart retrieval built in. Upload your data, call the API, and get accurate results from day one.

Get started freeRead the docs 
import { Agentset } from "agentset";

const agentset = new Agentset();
const ns = agentset.namespace("ns_1234");

const results = await ns.search(
  "What is multi-head attention?"
);

for (const result of results) {
  console.log(result.text);
}

Dataset Performance

By field

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

FiQa

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.843
0.804
Ranking quality at top 5 results
nDCG@10
0.835
0.812
Ranking quality at top 10 results
Recall@5
0.763
0.624
% of relevant docs in top 5
Recall@10
0.816
0.696
% of relevant docs in top 10
Latency Metrics
Mean
466ms
7ms
Average response time
P50
454ms
7ms
50th percentile (median)
P90
605ms
7ms
90th percentile

MSMARCO

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.956
0.952
Ranking quality at top 5 results
nDCG@10
0.939
0.941
Ranking quality at top 10 results
Recall@5
0.122
0.121
% of relevant docs in top 5
Recall@10
0.221
0.218
% of relevant docs in top 10
Latency Metrics
Mean
441ms
8ms
Average response time
P50
446ms
8ms
50th percentile (median)
P90
584ms
8ms
90th percentile

SciFact

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.871
0.696
Ranking quality at top 5 results
nDCG@10
0.871
0.702
Ranking quality at top 10 results
Recall@5
0.959
0.804
% of relevant docs in top 5
Recall@10
0.959
0.830
% of relevant docs in top 10
Latency Metrics
Mean
404ms
7ms
Average response time
P50
360ms
7ms
50th percentile (median)
P90
537ms
7ms
90th percentile

DBPedia

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.788
0.786
Ranking quality at top 5 results
nDCG@10
0.792
0.783
Ranking quality at top 10 results
Recall@5
0.061
0.061
% of relevant docs in top 5
Recall@10
0.120
0.122
% of relevant docs in top 10
Latency Metrics
Mean
436ms
7ms
Average response time
P50
432ms
7ms
50th percentile (median)
P90
592ms
7ms
90th percentile

business reports

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.091
0.000
Ranking quality at top 5 results
nDCG@10
0.084
0.000
Ranking quality at top 10 results
Recall@5
0.012
0.000
% of relevant docs in top 5
Recall@10
0.020
0.000
% of relevant docs in top 10
Latency Metrics
Mean
439ms
8ms
Average response time
P50
431ms
8ms
50th percentile (median)
P90
603ms
8ms
90th percentile

ARCD

MetricGemini Embedding 2Cohere Embed Multilingual v3Description
Accuracy Metrics
nDCG@5
0.868
0.868
Ranking quality at top 5 results
nDCG@10
0.875
0.875
Ranking quality at top 10 results
Recall@5
0.940
0.940
% of relevant docs in top 5
Recall@10
0.960
0.960
% of relevant docs in top 10
Latency Metrics
Mean
410ms
7ms
Average response time
P50
359ms
7ms
50th percentile (median)
P90
586ms
7ms
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.

View Full LeaderboardAll Leaderboards