GACL - Battleship — leaderboard
Game Agent Coding League: Battleship sub-benchmark. LLMs write code agents to play Battleship against each other.
Metric: Normalized Score (0-100). Source: gameagentcodingleague.com. Status: saturation imminent. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 Mini | 85.94 |
| 2 | GPT-5.5 | 85.38 |
| 3 | GPT-5.4 | 85 |
| 4 | Kimi K2.6 | 78.47 |
| 5 | GLM-5.1 | 71.19 |
| 6 | DeepSeek V4 Pro | 70.94 |
| 7 | Claude Opus 4.7 | 64.03 |
| 8 | Gemini 3.1 Pro (Preview) | 61.62 |
| 9 | DeepSeek V4 Flash | 61.41 |
| 10 | MiMo-V2.5-Pro | 59.44 |
| 11 | Claude Sonnet 4.6 | 59.25 |
| 12 | Claude Opus 4.6 | 58.81 |
| 13 | GPT-OSS-120B | 54.97 |
| 14 | Gemini 3 Flash (Preview) | 49.5 |
| 15 | Qwen 3.6 Max Preview | 41.44 |
Interactive version: theaggregate.ai/benchmark?slug=gacl-battleship · How the rankings work · Data refreshed daily, snapshot 2026-07-22.