Back to all rerankers

Contextual AI Rerank v2 Instruct

Available in 1B, 2B, and 6B parameter sizes with unique recency-awareness capabilities for time-sensitive ranking. Only reranker family capable of ranking recent information higher with ~35% performance improvement on recency tasks. If you want to compare the best rerankers for your data, try Agentset.

Leaderboard Rank
#9
of 12
ELO Rating
1469
#9
Win Rate
42.3%
#9
Accuracy (nDCG@10)
0.114
#1
Latency
3333ms
#11

Model Information

Provider
Contextual AI
License
cc-by-nc-4.0
Price per 1M tokens
$0.050
Release Date
2025-09-12
Model Name
ctxl-rerank-v2-instruct-multilingual
Total Evaluations
3300

Performance Record

Wins1397 (42.3%)
Losses1807 (54.8%)
Ties96 (2.9%)
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

Contextual AI Rerank v2 Instruct's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.

Contextual AI Rerank v2 Instruct - ELO by Dataset

Detailed Metrics

Dataset breakdown

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

business reports

ELO 152146.0% WR253W-295L-2T

Accuracy Metrics

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

Latency Distribution

Mean
3231ms
P50 (Median)
3129ms
P90
3651ms

PG

ELO 150656.7% WR312W-238L-0T

Accuracy Metrics

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

Latency Distribution

Mean
3566ms
P50 (Median)
3475ms
P90
4148ms

FiQa

ELO 147931.3% WR172W-364L-14T

Accuracy Metrics

nDCG@5
0.119
nDCG@10
0.125
Recall@5
0.123
Recall@10
0.135

Latency Distribution

Mean
3283ms
P50 (Median)
3209ms
P90
3891ms

MSMARCO

ELO 146346.0% WR253W-240L-57T

Accuracy Metrics

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

Latency Distribution

Mean
3283ms
P50 (Median)
3260ms
P90
3885ms

arguana

ELO 145239.6% WR218W-330L-2T

Accuracy Metrics

nDCG@5
0.525
nDCG@10
0.560
Recall@5
0.860
Recall@10
0.960

Latency Distribution

Mean
3627ms
P50 (Median)
3601ms
P90
4037ms

DBPedia

ELO 139634.4% WR189W-340L-21T

Accuracy Metrics

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

Latency Distribution

Mean
3010ms
P50 (Median)
3042ms
P90
3283ms

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 Contextual AI Rerank v2 Instruct with other top rerankers to understand the differences in performance, accuracy, and latency.