Claude Sonnet 5.5 (Medium): benchmark results
Provider: Anthropic. Released 2026-09-28. Access: API.
Aggregate Rating 1781 ± 15, rank #196 of 2286 rated models, from 69 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| The Aggregate - Lichess Pairwise Strength (25 Pairs) | 88 | Accuracy (%) | 94.4 |
| Conceptual Reasoning Index - Argument Evaluation (LMCA) | 65.45 | Chance-Corrected Score (0-100) | 94 |
| Conceptual Reasoning Index - Consistency (ACCoRD) | 83.39 | Chance-Corrected Score (0-100) | 92.7 |
| Conceptual Reasoning Index | 73.9 | Chance-Corrected Score (0-100) | 90.8 |
| AI for Education SEND | 84.4 | Accuracy (%) | 86.7 |
| AI for Education Pedagogy - Science | 91.26 | Accuracy (%) | 79.5 |
| AI for Education Pedagogy - Primary | 91.55 | Accuracy (%) | 77.9 |
| Maze-Bench | 58.47 | Mean Score | 77.1 |
| AI for Education Pedagogy - Secondary | 86.95 | Accuracy (%) | 76.4 |
| Conceptual Reasoning Index - Decision Theory (DTBench) | 89.78 | Chance-Corrected Score (0-100) | 75.3 |
| AI for Education Pedagogy | 87.54 | Accuracy (%) | 74.6 |
| o11y-bench - Pass@3 | 96.88 | Tasks passed on at least one of three attempts, Pass@3 (%) | 73.6 |
Interactive version: theaggregate.ai/model?slug=claude-sonnet-5-5-medium · How It Works · Data refreshed daily, snapshot 2026-11-20.