DuelLab Overall: leaderboard
DuelLab GameBench2 ranks AI-generated player programs across eight public board games. Published normalized scores combine relative conservative game ratings with a program-generation success penalty. Each primary provider reasoning configuration remains distinct; historical request variants and explicit thinking budgets are retained. Matches are repeated competitions between generated programs and are not independent generation trials.
Metric: GameBench2 normalized score (0–100). Source: benchmarks.duellab.org. Saturation forecast: Around January 2027. 208 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 (xHigh) | 95.17 | #3 (Claude Opus 5.5) |
| 2 | GPT-6 Astra (xHigh) | 94.53 | #6 (GPT-6 Astra) |
| 3 | Claude Opus 5.5 (Medium) | 92.82 | #3 (Claude Opus 5.5) |
| 4 | GPT-6.1 Sol (xHigh) | 92.68 | #7 (GPT-6.1 Sol) |
| 5 | Claude Sonnet 5.5 (xHigh) | 91.55 | #11 (Claude Sonnet 5.5) |
| 6 | GPT-6 Astra (Medium) | 88.97 | #6 (GPT-6 Astra) |
| 7 | GPT-6.1 Sol (Medium) | 87.23 | #7 (GPT-6.1 Sol) |
| 8 | Claude Opus 5 (xHigh) | 85.95 | #12 (Claude Opus 5) |
| 9 | GPT-6 Astra (Low) | 84.78 | #6 (GPT-6 Astra) |
| 10 | GPT-5.6 Sol (xHigh) | 81.31 | #19 (GPT-5.6 Sol) |
| 11 | GPT-6 Sol (xHigh) | 80.69 | #14 (GPT-6 Sol) |
| 12 | Claude Opus 5.5 (Low) | 77.82 | #3 (Claude Opus 5.5) |
| 13 | GPT-6.1 Sol (Low) | 77.36 | #7 (GPT-6.1 Sol) |
| 14 | Claude Opus 5 (Medium) | 76.65 | #12 (Claude Opus 5) |
| 15 | GPT-6 Sol (Medium) | 76.15 | #14 (GPT-6 Sol) |
No result here: #8 Gemini 4 Argon, #9 Claude Fable 5.1, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=duellab-overall · How It Works · Data refreshed daily, snapshot 2026-10-20.