A2Z GameSpec-Bench - GDD Fidelity: leaderboard
Metric: GDD Fidelity (%; coding-agent game design fidelity; Table 2). Source: a2z-gamespec-bench.github.io. Saturation forecast: Around April 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5.1 | 77 |
| 2 | Claude Opus 5 | 73.9 |
| 3 | GPT-6 Astra | 71.6 |
| 4 | GPT-5.5 | 63.2 |
| 5 | GPT-5.6 Sol | 61.8 |
| 6 | Claude Opus 4.8 | 56.8 |
| 7 | GLM-5.3 | 55.6 |
| 8 | DeepSeek V4 Pro | 50.4 |
Interactive version: theaggregate.ai/benchmark?slug=a2z-gamespec-bench-gdd-fidelity · How It Works · Data refreshed daily, snapshot 2026-10-10.