GeoR-Bench: leaderboard
Metric: Strict accuracy (%): share of samples whose output passes the reasoning, consistency and image-quality checks at once (mean of the six category accuracies), 440 reasoning-informed geoscience image-editing samples in six categories (geomorphology, hydrology, atmosphere and ocean, cryosphere, GIS and spatial geometry, crustal science), one generated output per sample, Gemini 3 Flash judge with task-specific rubrics; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-Image-2 | 42.7 |
| 2 | Nano Banana 2 | 39.2 |
| 3 | Nano Banana Pro | 35 |
| 4 | GPT-Image-1.5 | 26.1 |
| 5 | Seedream 5.0 | 19.6 |
| 6 | Nano Banana | 14.6 |
| 7 | Seedream 4.5 | 11.7 |
| 8 | FLUX 2 Max | 10.3 |
| 9 | Qwen-Image-Edit-2511 | 10.3 |
| 10 | FLUX 2 Pro | 5.63 |
Interactive version: theaggregate.ai/benchmark?slug=geor-bench · How It Works · Data refreshed daily, snapshot 2026-10-07.