Claude Opus 4.8 (High): benchmark results
Claude Opus 4.8 evaluated at the high reasoning-effort setting. Provider: Anthropic. Released 2026-05-28. Access: API.
Aggregate Rating 1815 ± 10, rank #110 of 2168 rated models, from 181 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CausalDS | 0.28 | CausalDS score (0 is perfect, lower is better, unbounded abo | 100 |
| FrontierFinance - Web Search Baseline | 32.98 | Rubric qualification rate (%, macro average over queries, th | 100 |
| OpenCompass Reasoning - Common | 80.6 | Score (%) | 100 |
| SWE-Interact | 26.7 | Resolve rate (%) of 75 multi-turn software-engineering tasks | 100 |
| SuperCLUE General (May 2026) - Code Generation | 83.58 | Score | 100 |
| SuperCLUE General (May 2026) - Hallucination Control | 87.48 | Score | 100 |
| SuperCLUE General (May 2026) - Science Reasoning | 77.19 | Score | 100 |
| SuperCLUE-SWE - Overall | 73.33 | Score | 100 |
| Horangi 4 - KMMLU-Pro | 95 | Accuracy (%) | 99.5 |
| o11y-bench - Pass^3 | 79.37 | Tasks passed on all three attempts, Pass^3 (%) | 99.1 |
| Horangi 4 - BigCodeBench (100-task subset) | 62 | Pass rate (%) | 98.1 |
| FutureEval | 14.12 | Unified Forecasting Score | 97.9 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-8-high · How It Works · Data refreshed daily, snapshot 2026-10-20.