Executable Games - Inter-Game Boundary Control: leaderboard
Executable Games (Fan et al., 2026) played after another game's full interaction history, without resetting the context, so the model must notice the new game and set the earlier evidence aside. Percent of games solved; higher is better.
Metric: Success Rate (%; each game starts after another game's full interaction history). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 77.47 | #91 |
| 2 | GPT-5.4 | 49.66 | #121 |
| 3 | DeepSeek V3.2 | 45.84 | #403 |
| 4 | Qwen 3 Max | 44.39 | #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-inter-game-boundary-control · How It Works · Data refreshed daily, snapshot 2026-11-12.