Back to all rerankers

Qwen3 Reranker 8B

State-of-the-art 8B parameter cross-encoder reranker built on Qwen3 foundation model. Achieves 69.02 on MTEB-R and 81.22 on MTEB-Code benchmarks. Supports 32K context, 100+ languages, and instruction-following capabilities for task-specific optimization. If you want to compare the best rerankers for your data, try Agentset.

Leaderboard Rank
#8
of 12
ELO Rating
1473
#8
Win Rate
51.2%
#7
Accuracy (nDCG@10)
0.106
#3
Latency
4687ms
#12

Model Information

Provider
Qwen
License
Open Source
Price per 1M tokens
$0.050
Release Date
2025-06-06
Model Name
qwen3-8b
Total Evaluations
3300

Performance Record

Wins1688 (51.2%)
Losses1501 (45.5%)
Ties111 (3.4%)
Wins
Losses
Ties

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

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

Performance Overview

ELO ratings by dataset

Qwen3 Reranker 8B's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.

Qwen3 Reranker 8B - ELO by Dataset

Detailed Metrics

Dataset breakdown

Performance metrics across different benchmark datasets, including accuracy and latency percentiles.

DBPedia

ELO 164455.5% WR305W-211L-34T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
1673ms
P50 (Median)
1673ms
P90
1787ms

FiQa

ELO 159360.0% WR330W-207L-13T

Accuracy Metrics

nDCG@5
0.114
nDCG@10
0.118
Recall@5
0.105
Recall@10
0.110

Latency Distribution

Mean
7242ms
P50 (Median)
2278ms
P90
2890ms

PG

ELO 158761.3% WR337W-213L-0T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
2567ms
P50 (Median)
2579ms
P90
2634ms

MSMARCO

ELO 153048.2% WR265W-231L-54T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
1728ms
P50 (Median)
1624ms
P90
1679ms

business reports

ELO 140145.1% WR248W-297L-5T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
1803ms
P50 (Median)
1763ms
P90
2097ms

arguana

ELO 108036.9% WR203W-342L-5T

Accuracy Metrics

nDCG@5
0.492
nDCG@10
0.519
Recall@5
0.800
Recall@10
0.880

Latency Distribution

Mean
13109ms
P50 (Median)
2812ms
P90
3425ms

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);
}

Compare Models

See how it stacks up

Compare Qwen3 Reranker 8B with other top rerankers to understand the differences in performance, accuracy, and latency.