GACL - Surround Morris: leaderboard
Game Agent Coding League: Surround Morris sub-benchmark. LLMs write code agents to play a mill/morris board game variant.
Metric: Normalized tournament points (0-100); source-selected best generated agent. Source: gameagentcodingleague.com. Saturation forecast: Around January 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 79.27 | #31 |
| 2 | DeepSeek V4 Pro | 75.47 | #102 |
| 3 | Claude Opus 4.7 | 68.44 | #49 |
| 4 | GPT-5.4 | 68.07 | #79 |
| 5 | Kimi K2.6 | 66.46 | #104 |
| 6 | Qwen 3.6 Plus | 63.23 | #120 |
| 7 | Gemini 3.1 Pro (Preview) | 60.73 | #58 |
| 8 | Claude Opus 4.6 | 53.7 | #63 |
| 9 | Claude Sonnet 4.6 | 51.67 | #90 |
| 10 | GLM-5.1 | 48.85 | #97 |
| 11 | MiniMax-M2.7 | 48.33 | #258 |
| 12 | Gemma 4 31B (IT) | 48.18 | #176 |
| 13 | DeepSeek V4 Flash | 42.4 | #137 |
| 14 | Qwen 3.6 Max Preview | 41.61 | #85 |
| 15 | MiMo-V2.5-Pro | 37.86 | #158 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gacl-surround-morris · How It Works · Data refreshed daily, snapshot 2026-10-20.