Qwen3 Reranker 8B vs Cohere Rerank 3.5
Detailed comparison between Qwen3 Reranker 8B and Cohere Rerank 3.5. See which reranker best meets your accuracy and performance needs. If you want to compare these models on your data, try Agentset.
Model Comparison
Qwen3 Reranker 8B takes the lead.
Both Qwen3 Reranker 8B and Cohere Rerank 3.5 are powerful reranking models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.
Why Qwen3 Reranker 8B:
- Qwen3 Reranker 8B has 21 higher ELO rating
- Qwen3 Reranker 8B delivers better accuracy (nDCG@10: 0.106 vs 0.080)
- Qwen3 Reranker 8B has a 10.3% higher win rate
Overview
Key metrics
ELO Rating
Overall ranking quality
Qwen3 Reranker 8B
Cohere Rerank 3.5
Win Rate
Head-to-head performance
Qwen3 Reranker 8B
Cohere Rerank 3.5
Accuracy (nDCG@10)
Ranking quality metric
Qwen3 Reranker 8B
Cohere Rerank 3.5
Average Latency
Response time
Qwen3 Reranker 8B
Cohere Rerank 3.5
Rerankers 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 reranking to configure.
Trusted by teams building production RAG applications
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
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Overall Performance | |||
| ELO Rating | 1473 | 1451 | Overall ranking quality based on pairwise comparisons |
| Win Rate | 51.2% | 40.9% | Percentage of comparisons won against other models |
| Pricing & Availability | |||
| Price per 1M tokens | $0.050 | $0.050 | Cost per million tokens processed |
| Release Date | 2025-06-06 | 2024-12-02 | Model release date |
| Accuracy Metrics | |||
| Avg nDCG@10 | 0.106 | 0.080 | Normalized discounted cumulative gain at position 10 |
| Performance Metrics | |||
| Avg Latency | 4687ms | 392ms | Average response time across all datasets |
Build RAG in Minutes, Not Months
Agentset gives you a complete RAG API with top-ranked rerankers and embedding models built in. Upload your data, call the API, and get accurate results from day one.
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.
MSMARCO
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Latency Metrics | |||
| Mean | 1728ms | 339ms | Average response time |
| P50 | 1624ms | 285ms | 50th percentile (median) |
| P90 | 1679ms | 304ms | 90th percentile |
arguana
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.492 | 0.267 | Ranking quality at top 5 results |
| nDCG@10 | 0.519 | 0.355 | Ranking quality at top 10 results |
| Recall@5 | 0.800 | 0.520 | % of relevant docs in top 5 |
| Recall@10 | 0.880 | 0.800 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 13109ms | 570ms | Average response time |
| P50 | 2812ms | 373ms | 50th percentile (median) |
| P90 | 3425ms | 617ms | 90th percentile |
FiQa
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.114 | 0.124 | Ranking quality at top 5 results |
| nDCG@10 | 0.118 | 0.128 | Ranking quality at top 10 results |
| Recall@5 | 0.105 | 0.123 | % of relevant docs in top 5 |
| Recall@10 | 0.110 | 0.130 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 7242ms | 364ms | Average response time |
| P50 | 2278ms | 315ms | 50th percentile (median) |
| P90 | 2890ms | 401ms | 90th percentile |
business reports
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Latency Metrics | |||
| Mean | 1803ms | 334ms | Average response time |
| P50 | 1763ms | 293ms | 50th percentile (median) |
| P90 | 2097ms | 503ms | 90th percentile |
PG
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Latency Metrics | |||
| Mean | 2567ms | 458ms | Average response time |
| P50 | 2579ms | 360ms | 50th percentile (median) |
| P90 | 2634ms | 615ms | 90th percentile |
DBPedia
| Metric | Qwen3 Reranker 8B | Cohere Rerank 3.5 | Description |
|---|---|---|---|
| Latency Metrics | |||
| Mean | 1673ms | 286ms | Average response time |
| P50 | 1673ms | 279ms | 50th percentile (median) |
| P90 | 1787ms | 290ms | 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.