Executable Games - Intra-Game Noise Injection: leaderboard
Executable Games (Fan et al., 2026) with irrelevant sentences inserted into the rules and meaningless fragments into the environment's responses, while the evidence the game reveals stays the same. Percent of games solved; higher is better.
Metric: Success Rate (%; irrelevant text inserted into rules and responses). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.3 | #91 |
| 2 | DeepSeek V3.2 | 44.19 | #403 |
| 3 | GPT-5.4 | 42.84 | #121 |
| 4 | Qwen 3 Max | 40.34 | #401 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=executable-games-intra-game-noise-injection · How It Works · Data refreshed daily, snapshot 2026-11-12.