Agentset

1.9k

OpenAI text-embedding-3-small vs Kanon 2

Detailed comparison between OpenAI text-embedding-3-small and Kanon 2. See which embedding best meets your accuracy and performance needs. If you want to compare these models on your data, try Agentset.

Explore OpenAI text-embedding-3-smallExplore Kanon 2

Model Comparison

OpenAI text-embedding-3-small takes the lead.

Both OpenAI text-embedding-3-small and Kanon 2 are powerful embedding models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.

Why OpenAI text-embedding-3-small:

  • OpenAI text-embedding-3-small has 39 higher ELO rating
  • OpenAI text-embedding-3-small delivers better accuracy (nDCG@10: 0.689 vs 0.484)
  • OpenAI text-embedding-3-small is 235ms faster on average
  • OpenAI text-embedding-3-small has a 10.5% higher win rate

Overview

Key metrics

ELO Rating

Overall ranking quality

OpenAI text-embedding-3-small

1489

Kanon 2

1450

Win Rate

Head-to-head performance

OpenAI text-embedding-3-small

43.9%

Kanon 2

33.5%

Accuracy (nDCG@10)

Ranking quality metric

OpenAI text-embedding-3-small

0.689

Kanon 2

0.484

Average Latency

Response time

OpenAI text-embedding-3-small

15ms

Kanon 2

250ms

Embedding Models 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 embeddings to manage.

Get started freeSchedule demo

Trusted by teams building production RAG applications

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

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

MetricOpenAI text-embedding-3-smallKanon 2Description
Overall Performance
ELO Rating
1489
1450
Overall ranking quality based on pairwise comparisons
Win Rate
43.9%
33.5%
Percentage of comparisons won against other models
Pricing & Availability
Price per 1M tokens
$0.020
$0.350
Cost per million tokens processed
Dimensions
1536
1792
Vector embedding dimensions (lower is more efficient)
Release Date
2024-01-25
2025-10-16
Model release date
Accuracy Metrics
Avg nDCG@10
0.689
0.484
Normalized discounted cumulative gain at position 10
Performance Metrics
Avg Latency
15ms
250ms
Average response time across all datasets

Build RAG in Minutes, Not Months

Agentset gives you a complete RAG API with top-ranked embedding models and smart retrieval built in. Upload your data, call the API, and get accurate results from day one.

Get started freeRead the docs 
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.

business reports

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.000
0.000
Ranking quality at top 5 results
nDCG@10
0.000
0.000
Ranking quality at top 10 results
Recall@5
0.000
0.000
% of relevant docs in top 5
Recall@10
0.000
0.000
% of relevant docs in top 10
Latency Metrics
Mean
16ms
250ms
Average response time
P50
16ms
250ms
50th percentile (median)
P90
16ms
250ms
90th percentile

DBPedia

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.858
0.806
Ranking quality at top 5 results
nDCG@10
0.807
0.777
Ranking quality at top 10 results
Recall@5
0.062
0.062
% of relevant docs in top 5
Recall@10
0.123
0.120
% of relevant docs in top 10
Latency Metrics
Mean
9ms
250ms
Average response time
P50
9ms
250ms
50th percentile (median)
P90
9ms
250ms
90th percentile

FiQa

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.801
0.839
Ranking quality at top 5 results
nDCG@10
0.814
0.836
Ranking quality at top 10 results
Recall@5
0.624
0.689
% of relevant docs in top 5
Recall@10
0.682
0.763
% of relevant docs in top 10
Latency Metrics
Mean
16ms
250ms
Average response time
P50
16ms
250ms
50th percentile (median)
P90
16ms
250ms
90th percentile

SciFact

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.663
0.718
Ranking quality at top 5 results
nDCG@10
0.684
0.744
Ranking quality at top 10 results
Recall@5
0.774
0.772
% of relevant docs in top 5
Recall@10
0.840
0.861
% of relevant docs in top 10
Latency Metrics
Mean
17ms
250ms
Average response time
P50
17ms
250ms
50th percentile (median)
P90
17ms
250ms
90th percentile

MSMARCO

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.959
0.941
Ranking quality at top 5 results
nDCG@10
0.946
0.931
Ranking quality at top 10 results
Recall@5
0.122
0.117
% of relevant docs in top 5
Recall@10
0.212
0.223
% of relevant docs in top 10
Latency Metrics
Mean
20ms
250ms
Average response time
P50
20ms
250ms
50th percentile (median)
P90
20ms
250ms
90th percentile

ARCD

MetricOpenAI text-embedding-3-smallKanon 2Description
Accuracy Metrics
nDCG@5
0.786
0.009
Ranking quality at top 5 results
nDCG@10
0.793
0.009
Ranking quality at top 10 results
Recall@5
0.900
0.020
% of relevant docs in top 5
Recall@10
0.920
0.020
% of relevant docs in top 10
Latency Metrics
Mean
15ms
250ms
Average response time
P50
15ms
250ms
50th percentile (median)
P90
15ms
250ms
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.

View Full LeaderboardAll Leaderboards