PlayWorld: leaderboard
PlayWorld has a human agent play inside each interactive world model toward long-horizon objectives, then rates the resulting rollout 1-5 on geometry, interaction, insight and out-of-sight evolution; the overall column averages the four.
Metric: Overall Score (1-5). Source: huggingface.co. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Genie 3 | 2.12 | |
| 2 | HappyOyster | 1.92 | |
| 3 | LingBot-World2 | 1.82 | |
| 4 | LingBot-World | 1.78 | |
| 5 | HY-World2 | 1.61 | |
| 6 | SANA-WM | 1.48 | |
| 7 | Hunyuan-GameCraft | 1.42 | |
| 8 | HY-WorldPlay | 1.21 | |
| 9 | Matrix-Game-3.0 | 1.14 |
Interactive version: theaggregate.ai/benchmark?slug=playworld · How It Works · Data refreshed daily, snapshot 2026-10-20.