Anthropic hybrid reasoning model with 200K context. First Claude with extended thinking. Exceptional code (95/100) and reasoning (96/100). Best for agentic tasks.
Metrics
Performance
ECI Score141.679
Epoch Capabilities Index score
ECI Normalized89
Normalized ECI score (0-100)
Context Window200,000
Maximum input token capacity
Training Compute3.3e25 FLOP
Total compute used during training
Info
ProviderAnthropic
Organization that created the model
CountryUSA
Model development country
Model AccessibilityAPI access
Access method (API, download, etc.)
Open WeightsNo
Model weights publicly available
Open SourceNo
Open source availability
Epoch Model IDclaude-3-7-sonnet-20250219_64K
Epoch AI benchmark identifier
Score Breakdown
code
20%89
Code generation, understanding, and debugging
math
20%86
Mathematical reasoning and problem solving
reasoning
15%90
Multi-step logical reasoning
intelligence
30%89
Overall reasoning and task completion ability
instruction following
88
Compatibility(4)
Sources(2)
Developed by
AnthropicBenchmark Performance
Compare ModelsEpoch Capabilities Index
141.7/ 155
| Dimension | Score |
|---|---|
| Intelligence | 89.0 |
| Reasoning | 90.0 |
| Math | 86.0 |
| Code | 89.0 |
| Instruction Following | 88.0 |
| Overall | 89 |
Loading chart...
Data: Epoch AI (CC-BY 4.0)