FutureX - Level 1 (Basic): leaderboard
FutureX Level 1 (Basic tier): single-choice questions with fewer than four options about real events that resolve after the prediction, answered by LLM search agents. Official runs from the newest week that has them, so the event set changes weekly. Higher is better, 0-100 (%); weighted 10% in the FutureX overall score.
Metric: Level 1 score (latest official week, %). Source: futurex.live. Saturation forecast: Around February 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 (High) | 66.67 | #83 (GPT-5.4) |
| 2 | GPT-5.5 (High) | 66.67 | #33 (GPT-5.5) |
| 3 | Gemini 3.5 Flash | 62.5 | #66 |
| 4 | Seed 2.0 Lite | 62.5 | #161 |
| 5 | Gemini 3.1 Pro (Preview) (High) | 58.33 | #64 (Gemini 3.1 Pro (Preview)) |
| 6 | Grok 4.20 (Reasoning) | 58.33 | #179 (Grok 4.20) |
| 7 | DeepSeek V3.2 (Thinking) | 54.17 | #210 (DeepSeek V3.2) |
| 8 | Qwen 3.7 Max | 54.17 | #85 |
| 9 | MiniMax-M2.7 | 50 | #270 |
| 10 | Qwen 3.6 Max Preview | 50 | #91 |
| 11 | Seed 2.0 Pro | 41.67 | #122 |
| 12 | DeepSeek V4 Flash | 37.5 | #155 |
| 13 | MiniMax-M3 | 33.33 | #151 |
| 14 | Grok 4 | 8.33 | #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-1-basic · How It Works · Data refreshed daily, snapshot 2026-11-05.