Design Arena — gamedev
Caveat: Head-to-head human votes on generated designs/code, mirrored via OpenRouter.
| # | Model | Score | Observed | |
|---|---|---|---|---|
| 1 | Claude Fable 5Anthropic | 1396 | Jul 23, 2026 | |
| 2 | GLM-5.2Zhipu AI | 1350 | Jul 23, 2026 | |
| 3 | Claude Sonnet 5Anthropic | 1345 | Jul 23, 2026 | |
| 4 | GPT-5.5OpenAI | 1343 | Jul 23, 2026 | |
| 5 | Claude Opus 4.6Anthropic | 1337 | Jul 23, 2026 | |
| 6 | Claude Opus 4.7Anthropic | 1336 | Jul 23, 2026 | |
| 7 | Grok 4.5xAI | 1330 | Jul 23, 2026 | |
| 8 | GLM-5.1Zhipu AI | 1328 | Jul 23, 2026 | |
| 9 | Muse Spark 1.1Meta | 1327 | Jul 23, 2026 | |
| 10 | Gemini 3.5 FlashGoogle | 1318 | Jul 24, 2026 | |
| 11 | Claude Sonnet 4.6Anthropic | 1315 | Jul 23, 2026 | |
| 11 | Qwen3.7 MaxAlibaba | 1315 | Jul 23, 2026 | |
| 13 | Kimi K2.6Moonshot AI | 1307 | Jul 23, 2026 | |
| 14 | Claude Opus 4.8Anthropic | 1302 | Jul 23, 2026 | |
| 15 | GPT-5.4OpenAI | 1298 | Jul 23, 2026 | |
| 16 | GLM-5Zhipu AI | 1292 | Jul 23, 2026 | |
| 17 | DeepSeek V4 ProDeepSeek | 1288 | Jul 23, 2026 | |
| 17 | Claude Opus 4.5 (latest)Anthropic | 1288 | Jul 23, 2026 | |
| 19 | GLM-5V-TurboZhipu AI | 1280 | Jul 23, 2026 | |
| 19 | MiniMax-M3MiniMax (minimax.io) | 1280 | Jul 23, 2026 | |
| 21 | Qwen3.7 PlusAlibaba | 1276 | Jul 23, 2026 | |
| 22 | Kimi K2.7 CodeMoonshot AI | 1270 | Jul 23, 2026 | |
| 23 | MiniMax-M2.7MiniMax (minimax.io) | 1265 | Jul 23, 2026 | |
| 24 | Kimi K2.5Moonshot AI | 1264 | Jul 23, 2026 | |
| 25 | Qwen3.6 PlusAlibaba | 1262 | Jul 23, 2026 | |
| 26 | Gemini 3.1 Pro PreviewGoogle | 1258 | Jul 23, 2026 | |
| 27 | Grok 4.20 (Non-Reasoning)xAI | 1255 | Jul 23, 2026 | |
| 28 | DeepSeek V4 FlashDeepSeek | 1254 | Jul 23, 2026 | |
| 29 | GPT-5.2OpenAI | 1253 | Jul 23, 2026 | |
| 30 | GLM-4.7Zhipu AI | 1247 | Jul 23, 2026 | |
| 31 | GPT-5OpenAI | 1245 | Jul 23, 2026 | |
| 32 | GPT-5.1OpenAI | 1237 | Jul 23, 2026 | |
| 33 | Grok 4.3xAI | 1235 | Jul 23, 2026 | |
| 33 | MiniMax-M2.5MiniMax (minimax.io) | 1235 | Jul 23, 2026 | |
| 35 | Claude Opus 4.1 (latest)Anthropic | 1230 | Jul 23, 2026 | |
| 36 | Claude Sonnet 4.5 (latest)Anthropic | 1225 | Jul 23, 2026 | |
| 36 | Gemini 3 Flash PreviewGoogle | 1225 | Jul 23, 2026 | |
| 38 | GPT-5.3 CodexOpenAI | 1219 | Jul 23, 2026 | |
| 39 | GLM-4.6Zhipu AI | 1208 | Jul 23, 2026 | |
| 40 | GLM-4.5Zhipu AI | 1203 | Jul 24, 2026 | |
| 41 | Step 3.7 FlashStepFun (China) | 1199 | Jul 24, 2026 | |
| 42 | Qwen3.5 397B-A17BAlibaba | 1198 | Jul 23, 2026 | |
| 43 | GPT-5.1 CodexOpenAI | 1197 | Jul 23, 2026 | |
| 44 | GLM-4.7-FlashZhipu AI | 1193 | Jul 23, 2026 | |
| 44 | MiniMax-M2.1MiniMax (minimax.io) | 1193 | Jul 23, 2026 | |
| 46 | DeepSeek V3.2DeepSeek | 1191 | Jul 23, 2026 | |
| 47 | GPT-5 MiniOpenAI | 1187 | Jul 23, 2026 | |
| 48 | Nemotron 3 Ultra 550B A55BNvidia | 1186 | Jul 23, 2026 | |
| 49 | MiniMax-M2MiniMax (minimax.io) | 1176 | Jul 23, 2026 | |
| 50 | Gemini 2.5 ProGoogle | 1170 | Jul 24, 2026 | |
| 51 | Qwen3.5 PlusAlibaba | 1164 | Jul 23, 2026 | |
| 52 | DeepSeek R1DeepSeek | 1157 | Jul 23, 2026 | |
| 53 | Claude Haiku 4.5 (latest)Anthropic | 1155 | Jul 23, 2026 | |
| 54 | Qwen3 MaxAlibaba | 1153 | Jul 23, 2026 | |
| 54 | GLM-4.5-AirZhipu AI | 1153 | Jul 23, 2026 | |
| 56 | GPT-5.1 Codex miniOpenAI | 1151 | Jul 23, 2026 | |
| 57 | GPT-4.1OpenAI | 1138 | Jul 23, 2026 | |
| 58 | GPT-4.1 miniOpenAI | 1129 | Jul 23, 2026 | |
| 59 | Gemini 2.5 FlashGoogle | 1124 | Jul 23, 2026 | |
| 60 | DeepSeek ChatDeepSeek | 1112 | Jul 23, 2026 | |
| 61 | GPT-5 NanoOpenAI | 1104 | Jul 23, 2026 | |
| 62 | o3OpenAI | 1093 | Jul 23, 2026 | |
| 63 | Gemini 3.1 Flash Lite PreviewGoogle | 1090 | Jul 23, 2026 | |
| 64 | o4-miniOpenAI | 1061 | Jul 23, 2026 | |
| 65 | Mercury 2Inception | 1036 | Jul 23, 2026 | |
| 66 | Kimi K2 ThinkingMoonshot AI | 1032 | Jul 24, 2026 | |
| 67 | GPT-4.1 nanoOpenAI | 1029 | Jul 23, 2026 | |
| 68 | Qwen3 235B-A22BthinkingAlibaba | 1016 | Jul 23, 2026 | |
| 69 | GPT-4oOpenAI | 965 | Jul 23, 2026 |
Dot = elo on a 943–1418 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.