Land or Water? (Open Models, 10° Reference) - AUC: leaderboard

Metric: ROC AUC. Source: github.com. Saturation forecast: Rough model projection: around 2026. 4 models tracked.

Top models

#ModelScore
1Gemma 4 26B A4B (IT) (Q8_0) (Non-reasoning)0.76
2Qwen 3.5 9B (Q8_0) (Non-reasoning)0.71
3GPT-OSS-120B MXFP4 (Non-reasoning)0.67
4Qwen 2.5 7B Instruct (Q8_0)0.65

Interactive version: theaggregate.ai/benchmark?slug=land-or-water-open-models-10-reference-auc · How It Works · Data refreshed daily, snapshot 2026-10-09.