FormationEval, an open multiple-choice benchmark for petroleum geoscience: leaderboard
Metric: accuracy (rate). Source: benchmarklist.com. 72 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 1 |
| 2 | GLM-4.7 | 0.99 |
| 3 | Gemini 3 Flash (Preview) | 0.98 |
| 4 | Gemini 2.5 Pro | 0.98 |
| 5 | Grok 4.1 Fast | 0.98 |
| 6 | Kimi K2 (Thinking) | 0.97 |
| 7 | Claude Opus 4.5 | 0.97 |
| 8 | GPT-5 Mini (Medium) | 0.96 |
| 9 | DeepSeek R1 | 0.96 |
| 10 | Grok 4 Fast | 0.96 |
| 11 | GPT-5 Mini (High) | 0.96 |
| 12 | O4 Mini (High) | 0.95 |
| 13 | GPT-5 Mini (Low) | 0.95 |
| 14 | Gemini 2.5 Flash | 0.95 |
| 15 | Grok 3 Mini | 0.95 |
Interactive version: theaggregate.ai/benchmark?slug=formationeval-an-open-multiple-choice-benchmark-for-petroleum-geoscience · How It Works · Data refreshed daily, snapshot 2026-09-19.