Land or Water? (Open Models, 10° Degrees) - AUC: leaderboard
Metric: ROC AUC. Source: github.com. Saturation forecast: Rough model projection: around 2026. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | OLMo 3.1 32B Instruct (Q8_0) | 0.71 |
| 2 | Qwen 3 4B (Q8_0) (Non-reasoning) | 0.61 |
| 3 | OLMo 3 7B Instruct (Q8_0) | 0.61 |
| 4 | Qwen 3 32B (Q8_0) (Non-reasoning) | 0.6 |
| 5 | OLMo 3 1025 7B (Q8_0) | 0.59 |
| 6 | Qwen 3 14B (Q8_0) (Non-reasoning) | 0.58 |
Interactive version: theaggregate.ai/benchmark?slug=land-or-water-open-models-10-degrees-auc · How It Works · Data refreshed daily, snapshot 2026-10-09.