Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort): benchmark results
Claude Sonnet 4.6 evaluated in adaptive-reasoning mode at max effort. Provider: Anthropic. Released 2026-02-17. Access: API.
Unified ELO 1652 ± 1, rank #239 of 1761 rated models, from 49 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Terminal-Bench Hard | 53.03 | Accuracy (%) | 96.2 |
| AA Global-MMLU-Lite - Japanese | 92.58 | Accuracy (%) | 95.9 |
| AA Global-MMLU-Lite - Portuguese | 92.5 | Accuracy (%) | 95.9 |
| AA Global-MMLU-Lite - Bengali | 90.92 | Accuracy (%) | 95.5 |
| AA Global-MMLU-Lite - Italian | 92.92 | Accuracy (%) | 95.5 |
| UGI - Writing | 64.15 | Writing Score | 94.8 |
| UGI Leaderboard | 52.82 | UGI Score | 94.3 |
| AA Global-MMLU-Lite - Korean | 91.08 | Accuracy (%) | 93.7 |
| UGI - Natural Intelligence | 58.7 | NatInt Score | 93.3 |
| AA Global-MMLU-Lite | 90.48 | Accuracy (%) | 91.9 |
| AA Global-MMLU-Lite - Arabic | 90.17 | Accuracy (%) | 91.9 |
| AA Global-MMLU-Lite - Burmese | 88.17 | Accuracy (%) | 91.9 |
Interactive version: theaggregate.ai/model?slug=claude-sonnet-4-6-adaptive-reasoning-max-effort · How It Works · Data refreshed daily, snapshot 2026-09-05.