PortBench-QA - VaR Estimation: leaderboard
Metric: Item score (%; historical-simulation value at risk from a supplied return series, 50 test questions per model). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.7 Max (Thinking) | 85.9 |
| 2 | Qwen 3.6 Plus (Thinking) | 85.8 |
| 3 | GLM-5.1 (Thinking) | 85.5 |
| 4 | DeepSeek V4 Flash (Thinking) | 84.3 |
| 5 | DeepSeek V4 Pro (Thinking) | 83.7 |
| 6 | Qwen 3.6 35B A3B (Thinking) | 80.8 |
| 7 | Kimi K2.6 (Thinking) | 42.2 |
Interactive version: theaggregate.ai/benchmark?slug=portbench-qa-var-estimation · How It Works · Data refreshed daily, snapshot 2026-09-26.