FutureX - Level 2 (Wide Search): leaderboard
FutureX Level 2 (Wide Search tier): multi-choice questions about future events with several correct options; missing a true option halves the item score and choosing a wrong one zeroes it. Official runs from the newest week that has them, so the event set changes weekly. Higher is better, 0-100 (%); weighted 20% overall.
Metric: Level 2 score (latest official week, %). Source: futurex.live. Saturation forecast: Around September 2028. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MiniMax-M3 | 44.44 | #151 |
| 2 | MiniMax-M2.7 | 44.44 | #270 |
| 3 | Gemini 3.5 Flash | 44.44 | #66 |
| 4 | DeepSeek V3.2 (Thinking) | 44.44 | #210 (DeepSeek V3.2) |
| 5 | Qwen 3.7 Max | 44.44 | #85 |
| 6 | Gemini 3.1 Pro (Preview) (High) | 44.44 | #64 (Gemini 3.1 Pro (Preview)) |
| 7 | GPT-5.4 (High) | 44.44 | #83 (GPT-5.4) |
| 8 | Seed 2.0 Pro | 44.44 | #122 |
| 9 | GPT-5.5 (High) | 44.44 | #33 (GPT-5.5) |
| 10 | Seed 2.0 Lite | 44.44 | #161 |
| 11 | Qwen 3.6 Max Preview | 44.44 | #91 |
| 12 | Grok 4.20 (Reasoning) | 44.44 | #179 (Grok 4.20) |
| 13 | DeepSeek V4 Flash | 33.33 | #155 |
| 14 | Grok 4 | 0 | #189 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=futurex-level-2-wide-search · How It Works · Data refreshed daily, snapshot 2026-11-05.