Voyage 3 Large vs Gemini text-embedding-004

Detailed comparison between Voyage 3 Large and Gemini text-embedding-004. See which embedding best meets your accuracy and performance needs.

Model Comparison

Voyage 3 Large takes the lead.

Both Voyage 3 Large and Gemini text-embedding-004 are powerful embedding models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.

Why Voyage 3 Large:

  • Voyage 3 Large has 81 higher ELO rating
  • Voyage 3 Large delivers better accuracy (nDCG@10: 0.837 vs 0.585)
  • Gemini text-embedding-004 is 20377ms faster on average
  • Voyage 3 Large has a 24.6% higher win rate

Overview

Key metrics

ELO Rating

Overall ranking quality

Voyage 3 Large

1528

Gemini text-embedding-004

1447

Win Rate

Head-to-head performance

Voyage 3 Large

52.6%

Gemini text-embedding-004

28.0%

Accuracy (nDCG@10)

Ranking quality metric

Voyage 3 Large

0.837

Gemini text-embedding-004

0.585

Average Latency

Response time

Voyage 3 Large

63477ms

Gemini text-embedding-004

43100ms

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

MetricVoyage 3 LargeGemini text-embedding-004Description
Overall Performance
ELO Rating
1528
1447
Overall ranking quality based on pairwise comparisons
Win Rate
52.6%
28.0%
Percentage of comparisons won against other models
Pricing & Availability
Price per 1M tokens
$0.180
$0.020
Cost per million tokens processed
Release Date
2025-01-07
2024-05-14
Model release date
Accuracy Metrics
Avg nDCG@10
0.837
0.585
Normalized discounted cumulative gain at position 10
Performance Metrics
Avg Latency
63477ms
43100ms
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

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
Latency Metrics
Mean
93447ms
76115ms
Average response time
P50
91578ms
74593ms
50th percentile (median)
P90
107464ms
87532ms
90th percentile

business reports

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
Latency Metrics
Mean
21279ms
9184ms
Average response time
P50
20853ms
9000ms
50th percentile (median)
P90
24471ms
10562ms
90th percentile

DBPedia

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
nDCG@5
0.675
0.536
Ranking quality at top 5 results
nDCG@10
0.638
0.517
Ranking quality at top 10 results
Recall@5
0.255
0.200
% of relevant docs in top 5
Recall@10
0.362
0.304
% of relevant docs in top 10
Latency Metrics
Mean
97066ms
59127ms
Average response time
P50
95125ms
57944ms
50th percentile (median)
P90
111626ms
67996ms
90th percentile

FiQa

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
nDCG@5
0.753
0.613
Ranking quality at top 5 results
nDCG@10
0.780
0.649
Ranking quality at top 10 results
Recall@5
0.742
0.645
% of relevant docs in top 5
Recall@10
0.837
0.748
% of relevant docs in top 10
Latency Metrics
Mean
80239ms
62373ms
Average response time
P50
78634ms
61126ms
50th percentile (median)
P90
92275ms
71729ms
90th percentile

SciFact

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
nDCG@5
0.796
0.722
Ranking quality at top 5 results
nDCG@10
0.809
0.745
Ranking quality at top 10 results
Recall@5
0.840
0.797
% of relevant docs in top 5
Recall@10
0.880
0.860
% of relevant docs in top 10
Latency Metrics
Mean
92317ms
65774ms
Average response time
P50
90471ms
64459ms
50th percentile (median)
P90
106165ms
75640ms
90th percentile

MSMARCO

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
nDCG@5
0.997
0.979
Ranking quality at top 5 results
nDCG@10
0.998
0.977
Ranking quality at top 10 results
Recall@5
0.122
0.118
% of relevant docs in top 5
Recall@10
0.222
0.209
% of relevant docs in top 10
Latency Metrics
Mean
75502ms
62416ms
Average response time
P50
73992ms
61168ms
50th percentile (median)
P90
86827ms
71778ms
90th percentile

NorQuAD

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
Latency Metrics
Mean
11003ms
5693ms
Average response time
P50
10783ms
5579ms
50th percentile (median)
P90
12653ms
6547ms
90th percentile

ARCD

MetricVoyage 3 LargeGemini text-embedding-004Description
Accuracy Metrics
nDCG@5
0.953
0.030
Ranking quality at top 5 results
nDCG@10
0.960
0.036
Ranking quality at top 10 results
Recall@5
0.960
0.040
% of relevant docs in top 5
Recall@10
0.980
0.060
% of relevant docs in top 10
Latency Metrics
Mean
36963ms
4118ms
Average response time
P50
36224ms
4036ms
50th percentile (median)
P90
42507ms
4736ms
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.