Back to all embeddings

Voyage 3.5 Lite

Most cost-optimized variant at $0.02 per 1M tokens achieving retrieval quality within 0.3% of Cohere-v4 at 1/6 the cost. Supports quantization options including 32-bit, int8, and binary precision for storage efficiency. If you want to compare the best embedding models for your data, try Agentset.

Leaderboard Rank
#9
of 18
ELO Rating
1490
#9
Win Rate
44.2%
#12
Accuracy (nDCG@10)
0.703
#5
Latency
19ms
#7

Model Information

Provider
Voyage AI
License
Proprietary
Price per 1M tokens
$0.020
Dimensions
512
Release Date
2025-05-20
Model Name
voyage-3.5-lite
Total Evaluations
830

Performance Record

Wins367 (44.2%)
Losses395 (47.6%)
Ties68 (8.2%)
Wins
Losses
Ties

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.

Trusted by teams building production RAG applications

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

Performance Overview

ELO ratings by dataset

Voyage 3.5 Lite's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.

Voyage 3.5 Lite - ELO by Dataset

Detailed Metrics

Dataset breakdown

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

business reports

ELO 150035.0% WR56W-102L-2T

Accuracy Metrics

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

Latency Distribution

Mean
54ms
P50 (Median)
54ms
P90
54ms

DBPedia

ELO 150061.9% WR99W-45L-16T

Accuracy Metrics

nDCG@5
0.793
nDCG@10
0.787
Recall@5
0.061
Recall@10
0.120

Latency Distribution

Mean
7ms
P50 (Median)
7ms
P90
7ms

FiQa

ELO 150044.0% WR66W-77L-7T

Accuracy Metrics

nDCG@5
0.812
nDCG@10
0.796
Recall@5
0.718
Recall@10
0.796

Latency Distribution

Mean
12ms
P50 (Median)
12ms
P90
12ms

SciFact

ELO 150036.3% WR58W-93L-9T

Accuracy Metrics

nDCG@5
0.704
nDCG@10
0.726
Recall@5
0.774
Recall@10
0.850

Latency Distribution

Mean
9ms
P50 (Median)
9ms
P90
9ms

MSMARCO

ELO 150040.0% WR64W-69L-27T

Accuracy Metrics

nDCG@5
0.965
nDCG@10
0.944
Recall@5
0.123
Recall@10
0.223

Latency Distribution

Mean
15ms
P50 (Median)
15ms
P90
15ms

ARCD

ELO 150060.0% WR24W-9L-7T

Accuracy Metrics

nDCG@5
0.874
nDCG@10
0.874
Recall@5
0.980
Recall@10
0.980

Latency Distribution

Mean
18ms
P50 (Median)
18ms
P90
18ms

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.

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 Voyage 3.5 Lite with other top embeddings to understand the differences in performance, accuracy, and latency.