INS-ActBench - Practice: leaderboard
Metric: Numerical accuracy (%; 0.2 percent relative tolerance) on the 1,152 spreadsheet and R-code tasks (INS-Act-Practice); 12,050 questions from the exams of 16 actuarial associations; knowledge and case questions two-shot, practice tasks zero-shot; proprietary models with thinking enabled at high where supported. Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (High) | 69.2 |
| 2 | Gemini 3.1 Pro (Preview) | 67.9 |
| 3 | Claude Opus 4.7 (High) | 65.4 |
| 4 | DeepSeek V4 Pro (High) | 61.7 |
| 5 | Qwen 3.6 Plus | 57.4 |
| 6 | Kimi K2.6 | 47.2 |
| 7 | Qwen 3.5 35B A3B | 37.7 |
| 8 | Qwen 3 14B | 26 |
| 9 | Gemma 3 12B (IT) | 20.4 |
Interactive version: theaggregate.ai/benchmark?slug=ins-actbench-practice · How It Works · Data refreshed daily, snapshot 2026-09-29.