ForecastBench Baseline - Zero-Shot - Dataset: leaderboard
Metric: Dataset Brier Index (difficulty-adjusted, higher is better). Source: www.forecastbench.org. 80 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 61.3 |
| 2 | GPT-5 | 61.1 |
| 3 | GPT-5.5 | 60.7 |
| 4 | Gemini 3 Pro (Preview) | 60.7 |
| 5 | O4 Mini (2025-04-16) | 60.7 |
| 6 | Gemini 3.1 Pro (Preview) | 60.6 |
| 7 | Gemini 3.5 Flash | 60.6 |
| 8 | Gemini 2.5 Flash (Preview 04-17) | 60.5 |
| 9 | Grok 4.20 0309 (Reasoning) | 60.5 |
| 10 | O3 (2025-04-16) | 60.4 |
| 11 | Grok 4.1 Fast (Reasoning) | 60.2 |
| 12 | Grok 4 Fast (Reasoning) | 60.2 |
| 13 | Gemini 3.1 Pro (Preview) (High) | 60.2 |
| 14 | Claude Opus 4.1 (20250805) | 60.2 |
| 15 | Grok 4 | 60.1 |
Interactive version: theaggregate.ai/benchmark?slug=forecastbench-baseline-zero-shot-dataset · How It Works · Data refreshed daily, snapshot 2026-09-19.