Claude Sonnet 5 (High) — benchmark results
Claude Sonnet 5 evaluated at the high reasoning-effort setting. Provider: Anthropic. Released 2026-06-30. Access: API.
Unified ELO 1970 ± 24, rank #36 of 1776 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LisanBench | 0.4 | Mean Path Length / Current Maximum | 95.3 |
| Chatbot Arena (Code) | 1544 | Elo | 91.9 |
| Chatbot Arena (Text) | 1461 | Elo | 90 |
| Agent Arena - Praise vs Complaint | 16.88 | Praise vs Complaint (%) | 86.5 |
| WeirdML | 68.78 | Average Score | 86.1 |
| Chatbot Arena (Vision) | 1272 | Arena Score | 85.8 |
| Multi-turn Debate (Lechmazur) | 1618.5 | Bradley-Terry Rating | 85 |
| Epoch AI - ECI | 153.2 | ECI Score | 80 |
| Agent Arena | 8.66 | Net Improvement (%) | 78.4 |
| Agent Arena - Confirmed Success | 8.14 | Confirmed Success (%) | 73 |
| Chatbot Arena (Document) | 1471 | Elo | 71 |
| Opus Magnum Bench | 26.55 | Human-normalized score (%) | 66.7 |
Interactive version: theaggregate.ai/model?slug=claude-sonnet-5-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.