Gemini 2.5 Pro (Medium): benchmark results
Provider: Google. Released 2025-03-25. Access: API.
Unified ELO 1670 ± 1, rank #236 of 3078 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| IKP - T7 | 21.51 | Accuracy (%) | 97.1 |
| IKP - T3 | 99.45 | Accuracy (%) | 95.4 |
| IKP - T6 | 55.08 | Accuracy (%) | 94.3 |
| Incompressible Knowledge Probes (IKP) | 78.87 | Mean tier accuracy (%) | 93.9 |
| IKP - T4 | 94.15 | Accuracy (%) | 93.3 |
| IKP - T2 | 100 | Accuracy (%) | 92.6 |
| IKP - T5 | 82.87 | Accuracy (%) | 90.1 |
| MageBench | 1655 | Rating | 78.9 |
| ObviousBench | 96.53 | Answer pass³ (%) | 70 |
| MageBench S1 | 1616 | Rating | 50 |
| GIM | 0.03 | IRT ability (theta) | 37.8 |
| IKP - T1 | 99 | Accuracy (%) | 35.3 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-medium · How It Works · Data refreshed daily, snapshot 2026-09-19.