Foresight Bench: leaderboard
Foresight Bench (Aleatoric AI) — LLM forecasting on real-world open questions; models keep updating predictions as events unfold, scored by mean Brier on resolved questions (reported as 100 − Brier×100).
Metric: Forecast Skill (100 - Brier x 100). Source: benchmark.aleatoricai.com. Saturation forecast: Estimated already saturated. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 91.67 | #16 |
| 2 | Kimi K2.6 | 91.59 | #104 |
| 3 | Claude Opus 4.8 | 91.17 | #37 |
| 4 | GPT-5.5 | 90.12 | #31 |
| 5 | Claude Opus 4.6 | 89.17 | #63 |
| 6 | Gemini 3.5 Flash | 89.14 | #60 |
| 7 | Gemini 3.1 Pro (Preview) | 89.12 | #58 |
| 8 | Claude Opus 4.7 | 88.39 | #49 |
| 9 | Qwen 3 235B A22B | 87.68 | #313 |
| 10 | Kimi K2 (Thinking) | 87.25 | #243 (Kimi K2) |
| 11 | Grok 4.3 | 87.07 | #188 |
| 12 | GPT-OSS-120B | 82.55 | #341 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=foresight-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.