IG-Bench - IG-Exam - Inheritance Tracing: leaderboard
Metric: Exact accuracy (%) on the 313 inheritance-tracing instances (12 task types) of IG-Exam. Source: arxiv.org. Saturation forecast: Around 2029. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 25.7 |
| 2 | Gemini 3.1 Pro (Preview) | 24.6 |
| 3 | Qwen 3.6 Max Preview | 22.5 |
| 4 | Kimi K2 (Thinking) | 22.1 |
| 5 | Claude Opus 4.7 | 21.9 |
| 6 | DeepSeek V4 Pro | 20.6 |
| 7 | GLM-5.1 | 19.7 |
| 8 | MiniMax-M2.7 | 9.1 |
Interactive version: theaggregate.ai/benchmark?slug=ig-bench-ig-exam-inheritance-tracing · How It Works · Data refreshed daily, snapshot 2026-09-29.