SciPredict: leaderboard
SciPredict benchmarks LLMs on forecasting the outcomes of real scientific experiments across biology, chemistry, and physics. (Source: benchmarklist.com, self-reported.)
Metric: Score (imported). Source: benchmarklist.com. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gemini-3-pro-preview | 25.27 | |
| 2 | claude-opus-4-5-20251101 | 23.05 | |
| 3 | claude-sonnet-4-5-20250929 | 22.55 | |
| 4 | gemini-3-flash | 22.22 | |
| 5 | claude-opus-4-1-20250805 | 22.22 | |
| 6 | gpt-5.2-2025-12-11 | 20.58 | |
| 7 | o3-mini-2025-01-31 | 19.84 | |
| 8 | DeepSeek-V3 | 19.18 | |
| 9 | Llama-3.3-70b | 18.19 | |
| 10 | o3-2025-04-03 | 17.94 |
Interactive version: theaggregate.ai/benchmark?slug=scipredict · How It Works · Data refreshed daily, snapshot 2026-10-20.