Foresight Bench — leaderboard
Metric: Forecast Skill (100 - Brier x 100). Source: benchmark.aleatoricai.com. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 91.93 |
| 2 | Kimi K2.6 | 91.59 |
| 3 | Claude Opus 4.7 | 90.51 |
| 4 | Claude Opus 4.6 | 90.08 |
| 5 | Claude Fable 5 | 89.75 |
| 6 | GPT-5.5 | 89.57 |
| 7 | Gemini 3.5 Flash | 89.19 |
| 8 | Gemini 3.1 Pro (Preview) | 87.75 |
| 9 | Kimi K2 (Thinking) | 87.53 |
| 10 | Grok 4.3 | 85.55 |
| 11 | Qwen 3 235B A22B | 85.22 |
| 12 | GPT-OSS-120B | 82.65 |
Interactive version: theaggregate.ai/benchmark?slug=foresight-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.