GPT-5 Pro (2025-10-06): benchmark results
Provider: OpenAI. Released 2025-08-07. Access: API.
Unified ELO 1830 ± 26, rank #39 of 1639 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEAL - MultiNRC | 65.2 | Score | 97.7 |
| SEAL - VISTA | 52.39 | Score | 95.2 |
| PLCC - Vocabulary | 92 | Accuracy (%) | 95.1 |
| PLCC - Culture & Tradition | 94 | Accuracy (%) | 92.8 |
| PM-LLM-Benchmark | 36.9 | Score | 91.9 |
| PLCC - Art & Entertainment | 88 | Accuracy (%) | 91.1 |
| PLCC - Overall | 91 | Mean category accuracy (%) | 90.7 |
| PLCC - Geography | 96 | Accuracy (%) | 89 |
| EnigmaEval | 18.75 | Score (self-reported) | 86.3 |
| SEAL - Humanity's Last Exam (Text Only) | 33.32 | Score | 85.7 |
| PLCC - Grammar | 85 | Accuracy (%) | 83.3 |
| PLCC - History | 91 | Accuracy (%) | 82 |
Interactive version: theaggregate.ai/model?slug=gpt-5-pro-2025-10-06 · How It Works · Data refreshed daily, snapshot 2026-10-09.