ForecastBench Tournament - Market: leaderboard
Market-question component of the ForecastBench tournament leaderboard: forecasts on questions from prediction markets and forecasting platforms, as a difficulty-adjusted Brier Index on 0-100 (higher is better; 50 equals always forecasting 0.5). The tournament accepts outside submissions with any scaffolding; ForecastBench-run models appear once each, with the market crowd forecast supplied where the run has that variant.
Metric: Market Brier Index (difficulty-adjusted, higher is better). Source: www.forecastbench.org. Saturation forecast: Around 2029. 363 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 (Max) | 71.3 | #20 (Claude Fable 5) |
| 2 | Claude Fable 5 (High) | 71 | #20 (Claude Fable 5) |
| 3 | Claude Fable 5.1 (Medium) | 70.9 | #10 (Claude Fable 5.1) |
| 4 | Gemini 3.7 Flash (High) | 70.8 | #24 (Gemini 3.7 Flash) |
| 5 | Grok 4 | 70.6 | #189 |
| 6 | Claude Opus 4.5 (20251101) | 70.6 | #81 |
| 7 | Gemini 3.6 Flash (Medium) | 70.5 | #57 (Gemini 3.6 Flash) |
| 8 | Gemini 3 Pro (Preview) | 70.4 | #71 |
| 9 | GPT-4.5 (Preview) | 70.4 | #260 |
| 10 | Grok Beta | 70.4 | #497 |
| 11 | Grok 4 Fast (Reasoning) | 70.3 | #266 (Grok 4 Fast) |
| 12 | Claude Sonnet 4 (20250514) | 70.2 | #223 |
| 13 | GPT-5 | 70 | #105 |
| 14 | Grok 4.1 Fast (Reasoning) | 69.9 | #227 (Grok 4.1 Fast) |
| 15 | Claude Opus 4.1 (20250805) | 69.7 | #139 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=forecastbench-tournament-market · How It Works · Data refreshed daily, snapshot 2026-11-05.