O1 (High) — benchmark results
O1 evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2024-12-05. Access: API.
Unified ELO 1682 ± 27, rank #304 of 1776 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PaperBench | 26 | Replication Score (%) | 100 |
| PlatinumBench (MIT) | 0.74 | Avg Error Rate (%) | 97 |
| ResearchCodeBench | 48.1 | Task Success Rate (%) | 74.2 |
| AI for Education Visual Maths - Geometry | 57.69 | Accuracy (%) | 72.5 |
| AI for Education Visual Maths | 61.6 | Accuracy (%) | 70.8 |
| AI for Education Visual Maths - Number and Operations | 54.05 | Accuracy (%) | 68.3 |
| AI for Education Visual Maths - Measurement | 81.08 | Accuracy (%) | 65.8 |
| AI for Education Visual Maths - Algebra | 76.92 | Accuracy (%) | 60.8 |
| Vals AI MATH 500 | 90.4 | Accuracy (%) | 58 |
| WeirdML | 46.08 | Average Score | 57.7 |
| Epoch AI - ECI | 142.71 | ECI Score | 48.6 |
| FutureEval | 8.5 | Unified Forecasting Score | 43.2 |
Interactive version: theaggregate.ai/model?slug=o1-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.