PortBench-QA - Position Sizing: leaderboard
Metric: Item score (%; fixed-fractional sizing from a drawdown limit and 99% VaR, 50 test questions per model). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 Plus (Thinking) | 96.8 |
| 2 | GLM-5.1 (Thinking) | 96.4 |
| 3 | DeepSeek V4 Pro (Thinking) | 96.3 |
| 4 | Qwen 3.6 35B A3B (Thinking) | 96.1 |
| 5 | Qwen 3.7 Max (Thinking) | 95.1 |
| 6 | DeepSeek V4 Flash (Thinking) | 94.5 |
| 7 | Kimi K2.6 (Thinking) | 49.3 |
Interactive version: theaggregate.ai/benchmark?slug=portbench-qa-position-sizing · How It Works · Data refreshed daily, snapshot 2026-09-26.