Competitive model from Google (ECI 131). solid benchmark performance. solid intelligence. Released Mar 2025. Open weights available.
Metrics
Performance
ECI Score130.946
Epoch Capabilities Index score
ECI Normalized81
Normalized ECI score (0-100)
Training Compute2.3e24 FLOP
Total compute used during training
Info
ProviderGoogle
Organization that created the model
CountryUK, USA
Model development country
Model AccessibilityOpen weights (restricted use)
Access method (API, download, etc.)
Open WeightsYes
Model weights publicly available
Epoch Model IDgemma-3-27b-it
Epoch AI benchmark identifier
Other
Release DateMar 12, 2025
Score Breakdown
code
20%69
Code generation, understanding, and debugging
math
20%73
Mathematical reasoning and problem solving
reasoning
15%77
Multi-step logical reasoning
intelligence
30%81
Overall reasoning and task completion ability
instruction following
73
Compatibility(6)
Sources(1)
Developed by
Google DeepMindBenchmark Performance
Compare ModelsEpoch Capabilities Index
130.9/ 155
| Dimension | Score |
|---|---|
| Intelligence | 81.0 |
| Reasoning | 77.0 |
| Math | 73.0 |
| Code | 69.0 |
| Instruction Following | 73.0 |
| Overall | 81 |
Loading chart...
Data: Epoch AI (CC-BY 4.0)