Gemini 2.5 Pro (Medium): benchmark results
Provider: Google. Released 2025-03-25. Access: API.
Aggregate Rating 1710 ± 19, rank #329 of 2170 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DALPHIN (Contextual) - Open Questions | 0.64 | BioBERT semantic similarity (biomedical sentence-embedding s | 100 |
| DALPHIN (Independent) - Multiple Choice | 0.84 | Accuracy (0-1; one case-specific multiple-choice question pe | 100 |
| DALPHIN (Independent) - Open Questions | 0.65 | BioBERT semantic similarity (biomedical sentence-embedding s | 100 |
| IKP - T7 | 21.51 | Accuracy (%) | 97.1 |
| CNSL-bench - Illustrative Images | 61.13 | Accuracy (%) identifying the sign's meaning from the diction | 95.5 |
| IKP - T3 | 99.45 | Accuracy (%) | 95.4 |
| CNSL-bench - Video (2 fps) | 48.32 | Accuracy (%) identifying the sign's meaning from its sign la | 95.2 |
| CNSL-bench - Text Descriptions | 84.79 | Accuracy (%) identifying the sign's meaning from the diction | 95 |
| CNSL-bench | 48.35 | Accuracy (%) identifying the sign's meaning from its sign la | 94.7 |
| 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 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-medium · How It Works · Data refreshed daily, snapshot 2026-11-05.