Executable Games - Intra-Game Noise Injection: leaderboard
Metric: Success Rate (%; irrelevant text inserted into rules and responses). Source: arxiv.org. Saturation forecast: Around December 2026. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.3 |
| 2 | DeepSeek V3.2 | 44.19 |
| 3 | GPT-5.4 | 42.84 |
| 4 | Qwen 3 Max | 40.34 |
Interactive version: theaggregate.ai/benchmark?slug=executable-games-intra-game-noise-injection · How It Works · Data refreshed daily, snapshot 2026-09-25.