Claude Sonnet 5 (Max): benchmark results
Claude Sonnet 5 evaluated at the max reasoning-effort setting. Provider: Anthropic. Released 2026-06-30. Access: API.
Aggregate Rating 1803 ± 10, rank #133 of 2168 rated models, from 157 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| IFM K2-Horizon-375B Card - APEX-Agents (Text-Only Subset) | 31.7 | Pass@1 (%) | 100 |
| IFM K2-Horizon-375B Card - AutomationBench (Public) | 34.7 | Score (%) | 100 |
| OmniaBench - DAG | 57.34 | Pass@1 (%; 354 multi-turn stateful tasks built from sampled | 100 |
| OmniaBench - ToB | 60.68 | Pass@1 (%; tasks from the business-facing (ToB) domain split | 100 |
| OmniaBench - ToC | 60.47 | Pass@1 (%; tasks from the consumer-facing (ToC) domain split | 100 |
| OmniaBench - Solver | 56.67 | Pass@1 (%; 60 selection, scheduling, allocation and optimiza | 95.2 |
| AA-Omniscience Index - Software Engineering (SWE) - Julia | 64 | Omniscience Index | 94 |
| AA Long Context Reasoning | 82 | Accuracy (%) | 91.9 |
| AA GPQA Diamond | 91.11 | Accuracy (%) | 91.7 |
| EnterpriseOps-Gym-AA - Teams | 49 | Task Success (%) | 90.8 |
| Vals AI TaxEval v2 | 75.63 | Accuracy (%) | 90.6 |
| MLCR-AA | 55 | Overall Score (%) | 90.5 |
Interactive version: theaggregate.ai/model?slug=claude-sonnet-5-max · How It Works · Data refreshed daily, snapshot 2026-10-20.