Skip to content
OneAtlas

Design Arena · Agentic game dev

Head-to-head Elo from Design Arena votes in the Agentic game dev category.

Published by Design Arena
24
Models scored
1249
Top score
1144
Median

Distribution of scores

9721249

Leaderboard

24 models

#ModelScoreWin rateContextInput $Output $
1Claude Fable 5Anthropic124961.9%1M$10.00$50.00
2Claude Opus 5Anthropic122553.4%1M$5.00$25.00
3Kimi K3Moonshot AI121855.0%1.05M$3.00$15.00
4Claude Opus 4.8Anthropic120355.5%1M$5.00$25.00
5Grok 4.5xAI120255.9%500K$2.00$6.00
6Gemini 3.7 FlashGoogle120052.0%1.05M$0.75$3.75
7Claude Sonnet 5Anthropic119051.8%1M$2.00$10.00
8Grok 4.6xAI117446.7%500K$2.00$6.00
9GLM-5.2Zhipu AI117149.1%1M$1.20$4.18
10Claude Opus 4.6Anthropic116446.3%1M$5.00$25.00
11Muse Spark 1.1Meta114748.1%1.05M$1.25$4.25
12Gemini 3.5 FlashGoogle114655.4%1.05M$1.50$9.00
13Gemini 3.6 FlashGoogle114447.5%1.05M$0.75$3.75
14GPT-5.5OpenAI114352.4%1.05M$5.00$30.00
15Muse Spark 1.2Meta114242.4%1.05M$1.25$4.25
16Claude Sonnet 4.6Anthropic114252.2%1M$3.00$15.00
17GLM-5.1Zhipu AI112849.5%200K$0.90$3.59
18MiniMax-M3MiniMax112544.1%1M$0.30$1.20
19Qwen3.7 MaxAlibaba112546.1%1M$2.50$7.50
20Kimi K2.6Moonshot AI110647.5%262K$0.95$4.00
21Kimi K2.7 CodeMoonshot AI109543.0%262K$1.90$8.00
22Gemini 3.1 Pro PreviewGoogle107944.1%1.05M$2.00$12.00
23GLM-5V-TurboZhipu AI106742.5%200K$0.75$3.29
24Grok 4.3xAI97228.7%1M$1.25$2.50