Voyage 3.5
Distilled from voyage-3-large for cost efficiency with flexible sizing options. Achieves 99% vector database cost reduction through binary quantization while maintaining strong retrieval quality.
Model Information
- Provider
- Voyage AI
- License
- Proprietary
- Price per 1M tokens
- $0.060
- Dimensions
- 1024
- Release Date
- 2025-05-20
- Model Name
- voyage-3.5
- Total Evaluations
- 1915
Performance Record
Wins934 (48.8%)
Losses738 (38.5%)
Ties243 (12.7%)
Wins
Losses
Ties
Performance Overview
ELO ratings by dataset
Voyage 3.5's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.
Voyage 3.5 - ELO by Dataset
Detailed Metrics
Dataset breakdown
Performance metrics across different benchmark datasets, including accuracy and latency percentiles.
ARCD
ELO 153848.8% WR117W-57L-66T
Accuracy Metrics
- nDCG@5
- 0.950
- nDCG@10
- 0.950
- Recall@5
- 0.980
- Recall@10
- 0.980
Latency Distribution
- Mean
- 23ms
- P50 (Median)
- 23ms
- P90
- 27ms
business reports
ELO 153358.3% WR140W-86L-14T
Latency Distribution
- Mean
- 16ms
- P50 (Median)
- 15ms
- P90
- 18ms
DBPedia
ELO 152858.3% WR137W-89L-9T
Accuracy Metrics
- nDCG@5
- 0.655
- nDCG@10
- 0.637
- Recall@5
- 0.246
- Recall@10
- 0.366
Latency Distribution
- Mean
- 6ms
- P50 (Median)
- 6ms
- P90
- 7ms
FiQa
ELO 151754.6% WR131W-107L-2T
Accuracy Metrics
- nDCG@5
- 0.721
- nDCG@10
- 0.741
- Recall@5
- 0.715
- Recall@10
- 0.793
Latency Distribution
- Mean
- 9ms
- P50 (Median)
- 9ms
- P90
- 11ms
NorQuAD
ELO 151230.4% WR73W-53L-114T
Latency Distribution
- Mean
- 20ms
- P50 (Median)
- 19ms
- P90
- 22ms
MSMARCO
ELO 150846.7% WR112W-102L-26T
Accuracy Metrics
- nDCG@5
- 1.000
- nDCG@10
- 1.000
- Recall@5
- 0.123
- Recall@10
- 0.224
Latency Distribution
- Mean
- 10ms
- P50 (Median)
- 9ms
- P90
- 11ms
SciFact
ELO 149546.3% WR111W-122L-7T
Accuracy Metrics
- nDCG@5
- 0.723
- nDCG@10
- 0.751
- Recall@5
- 0.778
- Recall@10
- 0.853
Latency Distribution
- Mean
- 14ms
- P50 (Median)
- 13ms
- P90
- 16ms
PG
ELO 149247.1% WR113W-122L-5T
Latency Distribution
- Mean
- 10ms
- P50 (Median)
- 10ms
- P90
- 12ms
Compare Models
See how it stacks up
Compare Voyage 3.5 with other top embeddings to understand the differences in performance, accuracy, and latency.