WorldCupBench - Brier Skill — leaderboard
Metric: 100 - Brier Total. Source: github.com. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 85.26 |
| 2 | MiMo-V2.5-Pro | 85.24 |
| 3 | MiniMax-M3 | 85.22 |
| 4 | Nex N2 Pro | 85.08 |
| 5 | Gemini 3.5 Flash | 85.05 |
| 6 | GPT-5.5 | 84.92 |
| 7 | Qwen 3.7 Max | 84.87 |
| 8 | Grok 4.3 | 84.82 |
| 9 | Kimi K2.6 | 84.76 |
| 10 | Claude Fable 5 | 84.65 |
| 11 | GLM-5.1 | 84.51 |
Interactive version: theaggregate.ai/benchmark?slug=worldcupbench-brier-skill · How the rankings work · Data refreshed daily, snapshot 2026-07-22.