AI Decathlon - Game Design & Development: leaderboard

AI Decathlon Timed Challenge (Game Design & Development) evaluating coding agents on 60-100 minute engineering, math, spatial, and interactive design tasks.

Metric: Game Design & Development Score (0-100). Source: octathlon.wendystars.com. Saturation forecast: Around February 2027. 25 models tracked. Updated 22 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 5.5xHigh88.5#9
2Claude Fable 5.2xHigh84.5#2
3GPT-6.2 AstraxHigh79.4#4
4Claude Sonnet 5.5xHigh78.8#24
5GPT-6.2 SolxHigh78.5#10
6Gemini 4 NeonxHigh75.4#3
7GPT-6 AstraxHigh74.1#11
8Gemini 4 CarbonHigh72.1#7
9Gemini 4 HydrogenxHigh71.2#16
10GPT-6.1 SolxHigh70.2#15
11Kimi K3.1 ProMax68.4#35
12GPT-6.1 TerraxHigh68#25
13Claude Haiku 5.5xHigh66.8#75
14DeepSeek V4.1 FlashMax65#69
15hy4High61.3#44

Interactive version: theaggregate.ai/benchmark?slug=ai-decathlon-game-design-development · How It Works · Data refreshed daily, snapshot 2026-11-22.