Claude 3.5 Sonnet (1022): benchmark results
Provider: Anthropic. Released 2024-10-22. Access: API.
Aggregate Rating 1594 ± 30, rank #1068 of 10916 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AGI-Eval Community - Cognition (English) | 82.07 | Accuracy (%) | 72.5 |
| AGI-Eval Community - Cognition | 76.72 | Accuracy (%) | 46.1 |
| AGI-Eval Community - Cognition (Chinese) | 72.84 | Accuracy (%) | 33.3 |
| AGI-Eval Community - Instruction Following | 32.75 | Accuracy (%) | 29.3 |
| AGI-Eval Community - Interaction (English) | 86.85 | Accuracy (%) | 26.4 |
| AGI-Eval Community - Algorithmic Reasoning | 19.15 | Accuracy (%) | 25 |
| AGI-Eval Community - Interaction (Chinese) | 64.22 | Accuracy (%) | 21 |
| AGI-Eval Community - Interaction | 70.16 | Accuracy (%) | 20.7 |
| AGI-Eval Community - Subject Reasoning (English) | 71.37 | Accuracy (%) | 19.6 |
| AGI-Eval Community - Objective Accuracy (English) | 79.86 | Accuracy (%) | 17.4 |
| AGI-Eval Community - Subject Reasoning | 65 | Accuracy (%) | 15.7 |
| AGI-Eval Community - General Reasoning | 90.43 | Accuracy (%) | 15.4 |
Interactive version: theaggregate.ai/model?slug=claude-3-5-sonnet-1022 · How It Works · Data refreshed daily, snapshot 2026-11-08.