GACL - Wizard — leaderboard
Game Agent Coding League: Wizard card game sub-benchmark. LLMs write code agents to play the trick-taking card game Wizard.
Metric: Normalized Score (0-100). Source: gameagentcodingleague.com. Status: saturation imminent. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 71.22 |
| 2 | GPT-5.4 | 66.07 |
| 3 | Gemini 3.1 Pro (Preview) | 61.77 |
| 4 | DeepSeek V4 Pro | 59.7 |
| 5 | Claude Opus 4.7 | 58.11 |
| 6 | Kimi K2.6 | 57.41 |
| 7 | Claude Opus 4.6 | 56.04 |
| 8 | Gemini 3 Flash (Preview) | 55.22 |
| 9 | Qwen 3.6 Plus | 55.15 |
| 10 | Qwen 3.6 Max Preview | 54.73 |
| 11 | MiMo-V2.5-Pro | 51.98 |
| 12 | GPT-OSS-120B | 48.69 |
| 13 | Gemma 4 31B (IT) | 47.92 |
| 14 | Claude Sonnet 4.6 | 47.61 |
| 15 | DeepSeek V4 Flash | 46.98 |
Interactive version: theaggregate.ai/benchmark?slug=gacl-wizard · How the rankings work · Data refreshed daily, snapshot 2026-07-22.