DuelLab GameBench 2: leaderboard
DuelLab GameBench 2 evaluates AI-generated game-playing programs by compiling and running them head-to-head on public board and strategy games, so the input modality is text/code. The pinned metric is Leaderboard score, measuring the released standings score reported for each system. This is a source-attributed result table mirrored by BenchmarkList; the evaluator, engine implementation, and unpublished operational artifacts remain private. Source: BenchmarkList’s attributed result table.
Metric: Leaderboard score. Source: benchmarklist.com. Saturation forecast: Around July 2027. 48 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 74.69 | #20 |
| 2 | Claude Opus 5 | 72.63 | #16 |
| 3 | GPT-5.6 Sol | 69.86 | #22 |
| 4 | GPT-5.4 | 57.05 | #83 |
| 5 | Grok 4.5 | 55.24 | #54 |
| 6 | Claude Opus 4.8 | 53.87 | #39 |
| 7 | GPT-5.5 | 53.27 | #33 |
| 8 | Gemini 3.6 Flash | 52.28 | #57 |
| 9 | Gemini 3.5 Flash | 51.31 | #66 |
| 10 | Kimi K3 | 49.99 | #35 |
| 11 | Muse Spark 1.2 | 48.33 | #69 |
| 12 | Claude Opus 4.5 | 48.11 | #87 |
| 13 | GPT-5.6 Terra | 47.63 | #40 |
| 14 | GLM-5.2 | 46.45 | #80 |
| 15 | Claude Sonnet 5 | 45.49 | #65 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=duellab-gamebench-2 · How It Works · Data refreshed daily, snapshot 2026-11-05.