Reference
Compare models
Select models and benchmarks. Scores are shown exactly within each benchmark’s own metric — never combined into an aggregate, average, or winner.
Models
1 selected
Claude Sonnet 4.6Anthropic
Benchmarks
2 selected
SWE-bench VerifiedCodingGPQA DiamondReasoning
| Benchmark | Claude Sonnet 4.6Anthropic |
|---|---|
| SWE-bench Verified% resolved | 75.2% |
| GPQA Diamondaccuracy | 87.4% |
SWE-bench Verified
% resolved
Claude Sonnet 4.6 · Anthropic
75.2%
GPQA Diamond
accuracy
Claude Sonnet 4.6 · Anthropic
87.4%
No aggregate score is calculated. Each row uses its benchmark’s own metric. Compare rows independently.