Cohere Embed v3 vs Qwen3 Embedding 0.6B
Detailed comparison between Cohere Embed v3 and Qwen3 Embedding 0.6B. See which embedding best meets your accuracy and performance needs.
Model Comparison
Cohere Embed v3 takes the lead.
Both Cohere Embed v3 and Qwen3 Embedding 0.6B are powerful embedding models designed to improve retrieval quality in RAG applications. However, their performance characteristics differ in important ways.
Why Cohere Embed v3:
- Cohere Embed v3 has 10 higher ELO rating
- Qwen3 Embedding 0.6B delivers better accuracy (nDCG@10: 0.751 vs 0.686)
Overview
Key metrics
ELO Rating
Overall ranking quality
Cohere Embed v3
Qwen3 Embedding 0.6B
Win Rate
Head-to-head performance
Cohere Embed v3
Qwen3 Embedding 0.6B
Accuracy (nDCG@10)
Ranking quality metric
Cohere Embed v3
Qwen3 Embedding 0.6B
Average Latency
Response time
Cohere Embed v3
Qwen3 Embedding 0.6B
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
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Overall Performance | |||
| ELO Rating | 1488 | 1478 | Overall ranking quality based on pairwise comparisons |
| Win Rate | 41.0% | 37.3% | Percentage of comparisons won against other models |
| Pricing & Availability | |||
| Price per 1M tokens | $0.100 | $0.010 | Cost per million tokens processed |
| Dimensions | 1024 | 1024 | Vector embedding dimensions (lower is more efficient) |
| Release Date | 2024-02-07 | 2025-06-06 | Model release date |
| Accuracy Metrics | |||
| Avg nDCG@10 | 0.686 | 0.751 | Normalized discounted cumulative gain at position 10 |
| Performance Metrics | |||
| Avg Latency | 7ms | 22ms | Average response time across all datasets |
Dataset Performance
By field
Comprehensive comparison of accuracy metrics (nDCG, Recall) and latency percentiles for each benchmark dataset.
PG
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| Latency Metrics | |||
| Mean | 6ms | 13ms | Average response time |
| P50 | 6ms | 13ms | 50th percentile (median) |
| P90 | 7ms | 15ms | 90th percentile |
business reports
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| Latency Metrics | |||
| Mean | 6ms | 22ms | Average response time |
| P50 | 6ms | 21ms | 50th percentile (median) |
| P90 | 7ms | 25ms | 90th percentile |
DBPedia
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.634 | 0.549 | Ranking quality at top 5 results |
| nDCG@10 | 0.619 | 0.556 | Ranking quality at top 10 results |
| Recall@5 | 0.219 | 0.216 | % of relevant docs in top 5 |
| Recall@10 | 0.353 | 0.350 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 6ms | 13ms | Average response time |
| P50 | 6ms | 13ms | 50th percentile (median) |
| P90 | 7ms | 15ms | 90th percentile |
FiQa
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.641 | 0.620 | Ranking quality at top 5 results |
| nDCG@10 | 0.650 | 0.647 | Ranking quality at top 10 results |
| Recall@5 | 0.639 | 0.590 | % of relevant docs in top 5 |
| Recall@10 | 0.678 | 0.680 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 7ms | 42ms | Average response time |
| P50 | 6ms | 41ms | 50th percentile (median) |
| P90 | 8ms | 49ms | 90th percentile |
SciFact
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.729 | 0.666 | Ranking quality at top 5 results |
| nDCG@10 | 0.769 | 0.686 | Ranking quality at top 10 results |
| Recall@5 | 0.788 | 0.723 | % of relevant docs in top 5 |
| Recall@10 | 0.900 | 0.783 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 8ms | 20ms | Average response time |
| P50 | 8ms | 19ms | 50th percentile (median) |
| P90 | 9ms | 23ms | 90th percentile |
MSMARCO
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 1.000 | 0.997 | Ranking quality at top 5 results |
| nDCG@10 | 0.996 | 0.992 | Ranking quality at top 10 results |
| Recall@5 | 0.123 | 0.122 | % of relevant docs in top 5 |
| Recall@10 | 0.218 | 0.215 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 6ms | 13ms | Average response time |
| P50 | 6ms | 13ms | 50th percentile (median) |
| P90 | 7ms | 15ms | 90th percentile |
NorQuAD
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| Latency Metrics | |||
| Mean | 9ms | 32ms | Average response time |
| P50 | 9ms | 31ms | 50th percentile (median) |
| P90 | 10ms | 36ms | 90th percentile |
ARCD
| Metric | Cohere Embed v3 | Qwen3 Embedding 0.6B | Description |
|---|---|---|---|
| Accuracy Metrics | |||
| nDCG@5 | 0.349 | 0.865 | Ranking quality at top 5 results |
| nDCG@10 | 0.398 | 0.872 | Ranking quality at top 10 results |
| Recall@5 | 0.380 | 0.880 | % of relevant docs in top 5 |
| Recall@10 | 0.520 | 0.900 | % of relevant docs in top 10 |
| Latency Metrics | |||
| Mean | 11ms | 23ms | Average response time |
| P50 | 11ms | 23ms | 50th percentile (median) |
| P90 | 12ms | 27ms | 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.