Gemini text-embedding-004
Supports 3,000 token context length with task-type specification for retrieval and classification. Legacy model scheduled for deprecation on January 14, 2026, replaced by gemini-embedding-001.
Model Information
- Provider
- License
- Proprietary
- Price per 1M tokens
- $0.020
- Dimensions
- 768
- Release Date
- 2024-05-14
- Model Name
- text-embedding-004
- Total Evaluations
- 1917
Performance Record
Wins536 (28.0%)
Losses1222 (63.7%)
Ties159 (8.3%)
Wins
Losses
Ties
Performance Overview
ELO ratings by dataset
Gemini text-embedding-004's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.
Gemini text-embedding-004 - ELO by Dataset
Detailed Metrics
Dataset breakdown
Performance metrics across different benchmark datasets, including accuracy and latency percentiles.
SciFact
ELO 148745.8% WR110W-126L-4T
Accuracy Metrics
- nDCG@5
- 0.722
- nDCG@10
- 0.745
- Recall@5
- 0.797
- Recall@10
- 0.860
Latency Distribution
- Mean
- 13ms
- P50 (Median)
- 13ms
- P90
- 15ms
DBPedia
ELO 147941.8% WR99W-133L-5T
Accuracy Metrics
- nDCG@5
- 0.536
- nDCG@10
- 0.517
- Recall@5
- 0.200
- Recall@10
- 0.304
Latency Distribution
- Mean
- 12ms
- P50 (Median)
- 11ms
- P90
- 13ms
NorQuAD
ELO 147917.9% WR43W-79L-118T
Latency Distribution
- Mean
- 13ms
- P50 (Median)
- 13ms
- P90
- 15ms
MSMARCO
ELO 147436.3% WR87W-132L-21T
Accuracy Metrics
- nDCG@5
- 0.979
- nDCG@10
- 0.977
- Recall@5
- 0.118
- Recall@10
- 0.209
Latency Distribution
- Mean
- 12ms
- P50 (Median)
- 12ms
- P90
- 14ms
FiQa
ELO 146237.1% WR89W-150L-1T
Accuracy Metrics
- nDCG@5
- 0.613
- nDCG@10
- 0.649
- Recall@5
- 0.645
- Recall@10
- 0.748
Latency Distribution
- Mean
- 12ms
- P50 (Median)
- 12ms
- P90
- 14ms
PG
ELO 142022.9% WR55W-184L-1T
Latency Distribution
- Mean
- 13ms
- P50 (Median)
- 13ms
- P90
- 15ms
business reports
ELO 141722.1% WR53W-186L-1T
Latency Distribution
- Mean
- 13ms
- P50 (Median)
- 12ms
- P90
- 14ms
ARCD
ELO 13580.0% WR0W-232L-8T
Accuracy Metrics
- nDCG@5
- 0.030
- nDCG@10
- 0.036
- Recall@5
- 0.040
- Recall@10
- 0.060
Latency Distribution
- Mean
- 13ms
- P50 (Median)
- 13ms
- P90
- 15ms
Compare Models
See how it stacks up
Compare Gemini text-embedding-004 with other top embeddings to understand the differences in performance, accuracy, and latency.