WorldCupBench
Unverified ML strategy on Multi by mverab. BotFinder score 18 out of 100.
⚽🤖 11 frontier LLMs predicted the entire 2026 World Cup — frozen before kickoff. Live leaderboard: Brier score, bracket points & Polymarket ROI.
Source: github
BotFinder analysis pending.
WorldCupBench
WorldCupBench ⚽🤖 The World Cup is the ultimate LLM eval. 10 frontier AI models predicted every match of the 2026 FIFA World Cup — frozen pre-tournament, scored live. 🇪🇸 Versión en Español --- 🏆 Live Leaderboard No predictions available yet. --- ⚡ How It Works (in 4 lines) 1. Same prompt → 10 SOTA LLMs via OpenRouter. 2. JSON predictions → every match, every round, every score, with 1X2 probabilities. 3. Frozen before kickoff → no post-hoc editing. Credibility is everything. 4. Scored live → as real results come in, we compute accuracy, Brier score, and ROI vs Polymarket. --- 🔮 Featured Predictions What do 10 frontier models agree on? What do they disagree on? Featured predictions will appear here once all models have submitted. Check back soon! --- 🚀 Quick Start The disagreement endpoint is available at http://localhost:8000/api/disagreement. --- 🤖 Compared Models (SOTA, June 2026) | Model | Provider | OpenRouter ID | |-------|----------|---------------| | GPT-5.5 | OpenAI | openai/gpt-5.5 | | Claude Fable 5 | Anthropic | anthropic/claude-fable-5 | | Gemini 3.5 Flash | Google | googl
⚠ No verified equity curve — no track-record source connected.
Drawdown profile
Data unavailable — contact the owner.
Verification ledger
How the score has moved
Recalculated at each data collection. Transparency means showing the bad weeks too.
No score history is stored yet — only the current score is shown.
Reviews & comments
No reviews collected from the source yet.
⚠ No live verification account connected — ask for proof before buying.
Alerts on changes: coming soon
Prop-firm compatibility not provided.
Open-source maintainer on GitHub.
Data-completeness & trust index (not a profitability rating)