Gemini 2.5 Pro (Preview): benchmark results
Google preview Gemini 2.5 Pro model, an early Pro snapshot from the Gemini 2.5 family. Provider: Google. Released 2025-03-25. Access: API.
Unified ELO 1830 ± 18, rank #106 of 2653 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MATH-Perturb (Hard) | 88.89 | Accuracy (%) | 100 |
| PlatinumBench (MIT) | 0.53 | Avg Error Rate (%) | 100 |
| LLMEval-Fair - Management | 9.63 | Discipline Score (0-10) | 98.3 |
| LLMEval-Fair - Military | 8.87 | Discipline Score (0-10) | 98.3 |
| IKP - T7 | 22.09 | Accuracy (%) | 98.2 |
| Gapminder AI Worldview | 93.1 | Correct Rate (%) | 97.1 |
| LLMEval-Fair - Overall | 91.07 | Absolute Score (0-100) | 96.6 |
| LLMEval-Fair - Science | 9.07 | Discipline Score (0-10) | 96.6 |
| IKP - T3 | 99.45 | Accuracy (%) | 95.4 |
| IKP - T4 | 95.21 | Accuracy (%) | 95.4 |
| IKP - T6 | 57.75 | Accuracy (%) | 95 |
| LLMEval-Fair - Economics | 9.47 | Discipline Score (0-10) | 94.8 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-preview · How It Works · Data refreshed daily, snapshot 2026-09-21.