GameWorld Generalist: leaderboard
Generalist multimodal agents playing 34 browser games (170 tasks). Measures overall progress percentage using state-verifiable evaluation across runner, arcade, platformer, puzzle, and simulation genres.
Metric: Progress (%). Source: gameworld-project.github.io. Saturation forecast: Around April 2028. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Median Expert | 82.6 | |
| 2 | Median Human | 64.1 | |
| 3 | Gemini 3 Flash (Preview) | 41.9 | #82 |
| 4 | GPT-5.2 | 40.6 | #110 |
| 5 | Claude Sonnet 4.6 | 39.3 | #90 |
| 6 | Seed 1.8 | 39 | #139 |
| 7 | Kimi K2.5 | 37.4 | #144 |
| 8 | Grok 4.1 Fast (Reasoning) | 36 | #217 (Grok 4.1 Fast) |
| 9 | Qwen 3 VL 235B A22B (Thinking) | 30.8 | #238 (Qwen 3 VL 235B A22B) |
| 10 | GLM-4.6V | 30.8 | #316 |
| 11 | Qwen 3 VL 30B A3B (Thinking) | 30.6 | #351 (Qwen 3 VL 30B A3B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gameworld-generalist · How It Works · Data refreshed daily, snapshot 2026-10-20.