Compare / Claude Sonnet 4.6 vs Claude Sonnet 5.5
Claude Sonnet 4.6 vs Claude Sonnet 5.5
7 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Sonnet 4.6 | Claude Sonnet 5.5 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 43.0% | 75.5% |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 30.1 | 56.0 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 33.6% | 55.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 80.0% | 82.7% |
| Artificial Analysisreported | SciCode% of problems | 50.1% | 61.0% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 3.0% | 63.6% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 35.5% | 46.5% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.