A2Z GameSpec-Bench - Source Code Small: leaderboard
Metric: Source Code Small (%; coding-agent game design fidelity; Table 2). Source: a2z-gamespec-bench.github.io. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5.1 | 89.7 |
| 2 | Claude Opus 5 | 87.9 |
| 3 | GPT-6 Astra | 84.2 |
| 4 | GPT-5.5 | 81 |
| 5 | GLM-5.3 | 78.6 |
| 6 | GPT-5.6 Sol | 78.5 |
| 7 | DeepSeek V4 Pro | 75.7 |
| 8 | Claude Opus 4.8 | 72.5 |
Interactive version: theaggregate.ai/benchmark?slug=a2z-gamespec-bench-source-code-small · How It Works · Data refreshed daily, snapshot 2026-10-10.