ForecastBench Tournament Dataset Brier Index
Forecasting · 2024-09-30
Difficulty-adjusted Brier Index on ForecastBench dataset questions.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.5 | 63.3 |
| Grok 4.20 | 63.0 |
| Gemini 3.7 Flash | 62.2 |
| Opus 4.8 | 61.6 |
| GPT-5.6 Sol | 61.6 |
| Claude Opus 5 | 61.5 |
| Gemini 3.6 Flash | 61.5 |
| Gemini 2.5 Pro Preview 03-25 | 61.3 |
| Gemini 3.1 Pro Preview | 61.2 |
| Grok 4.5 | 61.2 |
| GPT-4.5 | 61.1 |
| GPT-5 | 61.0 |
| Opus 4 | 60.8 |
| Grok 4.6 | 60.8 |
| Gemini 2.5 Flash Preview 04-17 | 60.7 |
| Gemini 3 Pro Preview | 60.6 |
| GPT-5.6 Sol Pro | 60.6 |
| O4 Mini | 60.6 |
| Gemini 3.5 Flash | 60.5 |
| GLM-5 | 60.4 |
| Kimi K3 | 60.4 |
| o3 | 60.3 |
| Gemini 3 Flash Preview | 60.2 |
| GPT-4.1 | 60.2 |
| Opus 4.1 | 60.1 |
| Grok 4.1 Fast | 60.1 |
| Grok 4 Fast | 60.1 |
| Claude 3.7 Sonnet | 60.1 |
| Sonnet 4.6 | 60.0 |
| GPT-5 Mini | 60.0 |
| Grok 4 | 60.0 |
| Opus 4.5 | 59.9 |
| Opus 4.7 | 59.9 |
| Gemini 3.1 Flash-Lite Preview | 59.9 |
| MiniMax M3 | 59.9 |
| Opus 4.6 | 59.8 |
| Sonnet 4.5 | 59.8 |
| Gemini 2.5 Pro | 59.8 |
| GPT-5.2 | 59.8 |
| Grok 4.3 | 59.8 |