GACL - 2x8 MiniChess: leaderboard
Game Agent Coding League: 2x8 MiniChess sub-benchmark. LLMs write code agents to play a simplified chess variant on a 2x8 board.
Metric: Normalized tournament points (0-100); source-selected best generated agent. Source: gameagentcodingleague.com. Saturation forecast: Around May 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 69.06 | #31 |
| 2 | GPT-5.4 | 67.4 | #79 |
| 3 | Claude Opus 4.6 | 60.21 | #63 |
| 4 | MiMo-V2.5-Pro | 52.08 | #158 |
| 5 | Gemini 3.1 Pro (Preview) | 50.83 | #58 |
| 6 | GLM-5.1 | 50.47 | #97 |
| 7 | GPT-5.4 Mini | 49.43 | #212 |
| 8 | Kimi K2.6 | 48.13 | #104 |
| 9 | Qwen 3.6 Max Preview | 45.47 | #85 |
| 10 | Claude Sonnet 4.6 | 43.8 | #90 |
| 11 | DeepSeek V4 Pro | 43.8 | #102 |
| 12 | Claude Opus 4.7 | 43.18 | #49 |
| 13 | Qwen 3.6 Plus | 40.42 | #120 |
| 14 | GPT-OSS-120B | 38.85 | #341 |
| 15 | Qwen 3.6 35B A3B | 38.13 | #226 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gacl-2x8-minichess · How It Works · Data refreshed daily, snapshot 2026-10-20.