Alpha Arena (Nof1 AI): leaderboard
Live trading benchmark giving LLMs real capital to trade crypto perpetuals autonomously. Measures actual P&L, risk management, and decision-making under real market conditions.
Metric: Return (%). Source: www.alphaarena.ai. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.20 | 34.59 | #175 |
| 2 | GPT-5.1 | 8.88 | #135 |
| 3 | DeepSeek V3.1 | 7.3 | #264 |
| 4 | Qwen 3 Max | -6.79 | #208 |
| 5 | Kimi K2 (Thinking) | -10.45 | #243 (Kimi K2) |
| 6 | Gemini 3 Pro | -10.94 | #80 |
| 7 | Claude Sonnet 4.5 | -19.15 | #142 |
| 8 | Grok 4 | -48.4 | #181 |
Interactive version: theaggregate.ai/benchmark?slug=alpha-arena-nof1-ai · How It Works · Data refreshed daily, snapshot 2026-10-20.