Index
sid_tppPAkJqjQ / MMLU-Pro: 89.5
Verdictconfirmed99%
1 check · 4/24/2026Inline sourcing: confirmed
Our claim
entire record- Benchmark
- 3PM0ZfVJxU
- Model
- Claude Opus 4.5
- Score
- 89.5
- Unit
- percent
- Date
- November 24, 2025
- Notes
- MMLU-Pro - enhanced MMLU with harder 10-option multiple choice format
Source evidence
1 src · 1 checkconfirmed99%inline-submission · 4/24/2026
Case № 0x0bVn7TBvFiled 4/24/2026Confidence 99%