Cohere Embed Multilingual v3
Supports 100+ languages with 512 token context length enabling within-language and cross-language retrieval. Requires input_type specification for search documents and queries to optimize multilingual search performance.
Model Information
- Provider
- Cohere
- License
- Proprietary
- Price per 1M tokens
- $0.100
- Dimensions
- 512
- Release Date
- 2024-02-07
- Model Name
- cohere-embed-multilingual-v3
- Total Evaluations
- 1920
Performance Record
Wins823 (42.9%)
Losses792 (41.3%)
Ties305 (15.9%)
Wins
Losses
Ties
Performance Overview
ELO ratings by dataset
Cohere Embed Multilingual v3's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.
Cohere Embed Multilingual v3 - ELO by Dataset
Detailed Metrics
Dataset breakdown
Performance metrics across different benchmark datasets, including accuracy and latency percentiles.
MSMARCO
ELO 151244.2% WR106W-87L-47T
Accuracy Metrics
- nDCG@5
- 0.996
- nDCG@10
- 0.994
- Recall@5
- 0.122
- Recall@10
- 0.218
Latency Distribution
- Mean
- 6ms
- P50 (Median)
- 6ms
- P90
- 7ms
NorQuAD
ELO 150521.3% WR51W-40L-149T
Latency Distribution
- Mean
- 10ms
- P50 (Median)
- 9ms
- P90
- 11ms
ARCD
ELO 150340.0% WR96W-82L-62T
Accuracy Metrics
- nDCG@5
- 0.925
- nDCG@10
- 0.933
- Recall@5
- 0.940
- Recall@10
- 0.960
Latency Distribution
- Mean
- 8ms
- P50 (Median)
- 7ms
- P90
- 9ms
business reports
ELO 150249.2% WR118W-110L-12T
Latency Distribution
- Mean
- 7ms
- P50 (Median)
- 7ms
- P90
- 8ms
SciFact
ELO 150149.6% WR119W-115L-6T
Accuracy Metrics
- nDCG@5
- 0.723
- nDCG@10
- 0.732
- Recall@5
- 0.808
- Recall@10
- 0.833
Latency Distribution
- Mean
- 8ms
- P50 (Median)
- 8ms
- P90
- 9ms
DBPedia
ELO 150047.9% WR115W-111L-14T
Accuracy Metrics
- nDCG@5
- 0.619
- nDCG@10
- 0.591
- Recall@5
- 0.222
- Recall@10
- 0.329
Latency Distribution
- Mean
- 7ms
- P50 (Median)
- 7ms
- P90
- 8ms
PG
ELO 149445.4% WR109W-120L-11T
Latency Distribution
- Mean
- 6ms
- P50 (Median)
- 6ms
- P90
- 7ms
FiQa
ELO 148945.4% WR109W-127L-4T
Accuracy Metrics
- nDCG@5
- 0.637
- nDCG@10
- 0.654
- Recall@5
- 0.621
- Recall@10
- 0.692
Latency Distribution
- Mean
- 7ms
- P50 (Median)
- 7ms
- P90
- 9ms
Compare Models
See how it stacks up
Compare Cohere Embed Multilingual v3 with other top embeddings to understand the differences in performance, accuracy, and latency.