Cohere Rerank 4 Pro vs Voyage AI Rerank 2.5 Lite

Detailed comparison between Cohere Rerank 4 Pro and Voyage AI Rerank 2.5 Lite. See which reranker best meets your accuracy and performance needs.

Model Comparison

Cohere Rerank 4 Pro takes the lead.

Both Cohere Rerank 4 Pro and Voyage AI Rerank 2.5 Lite are powerful reranking models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.

Why Cohere Rerank 4 Pro:

  • Cohere Rerank 4 Pro has 99 higher ELO rating
  • Cohere Rerank 4 Pro has a 5.1% higher win rate

Overview

Key metrics

ELO Rating

Overall ranking quality

Cohere Rerank 4 Pro

1627

Voyage AI Rerank 2.5 Lite

1528

Win Rate

Head-to-head performance

Cohere Rerank 4 Pro

58.2%

Voyage AI Rerank 2.5 Lite

53.1%

Accuracy (nDCG@10)

Ranking quality metric

Cohere Rerank 4 Pro

N/A

Voyage AI Rerank 2.5 Lite

0.103

Average Latency

Response time

Cohere Rerank 4 Pro

614ms

Voyage AI Rerank 2.5 Lite

616ms

Visual Performance Analysis

Performance

ELO Rating Comparison

Win/Loss/Tie Breakdown

Latency Distribution (ms)

Breakdown

How the models stack up

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Overall Performance
ELO Rating
1627
1528
Overall ranking quality based on pairwise comparisons
Win Rate
58.2%
53.1%
Percentage of comparisons won against other models
Pricing & Availability
Price per 1M tokens
$0.050
$0.020
Cost per million tokens processed
Release Date
2025-12-11
2025-08-11
Model release date
Accuracy Metrics
Avg nDCG@10
N/A
0.103
Normalized discounted cumulative gain at position 10
Performance Metrics
Avg Latency
614ms
616ms
Average response time across all datasets

Dataset Performance

By field

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

MSMARCO

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
458ms
563ms
Average response time
P50
408ms
610ms
50th percentile (median)
P90
615ms
619ms
90th percentile

arguana

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
785ms
636ms
Average response time
P50
768ms
613ms
50th percentile (median)
P90
933ms
819ms
90th percentile

FiQa

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
610ms
639ms
Average response time
P50
585ms
613ms
50th percentile (median)
P90
817ms
819ms
90th percentile

business reports

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
529ms
599ms
Average response time
P50
498ms
611ms
50th percentile (median)
P90
675ms
727ms
90th percentile

PG

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
760ms
670ms
Average response time
P50
720ms
615ms
50th percentile (median)
P90
896ms
818ms
90th percentile

DBPedia

MetricCohere Rerank 4 ProVoyage AI Rerank 2.5 LiteDescription
Latency Metrics
Mean
541ms
587ms
Average response time
P50
489ms
612ms
50th percentile (median)
P90
729ms
656ms
90th percentile

Explore More

Compare more rerankers

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