Design Arena: 3D leaderboard
- GPT-6 Astra leads
- 1500
- models measured
- 116
- lead over #2
- +24 Elo
- median
- 1189
Leader on Design Arena: 3D · read October 22
On October 22, 2026, 119 of our catalog's models had a Design Arena: 3D result, and OpenAI GPT-6 Astra led with 1500, ahead of Anthropic Claude Opus 5.5 at 1476.
Design Arena shows people two designs made by two models from the same prompt (a 3D scene) and asks which is better. The number is the Elo rating built from those votes, as Design Arena publishes it.
Role in the index
Design Arena: 3D is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Battles | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 1500 | 265 | 85.1 | $50.00 | |
| 2 | Claude Opus 5.5Anthropic | 1476 | 862 | 88.8 | $20.00 | |
| 3 | GPT-6.1 SolOpenAI | 1460 | 380 | 82.4 | $10.00 | |
| 4 | Claude Sonnet 5.5Anthropic | 1445 | 274 | 87.3 | $10.00 | |
| 5 | Muse WonderMeta | 1427 | 610 | 71.9 | $20.00 | |
| 6 | Claude Fable 5.1Anthropic | 1392 | 1,271 | 84.2 | $50.00 | |
| 7 | Gemini 4 ArgonGoogle | 1388 | 1,183 | 81.0 | $10.00 | |
| 8 | Kimi K3Moonshot AI | 1387 | 3,370 | 62.3 | $15.00 | |
| 9 | GPT-5.6 SolOpenAI | 1384 | 2,452 | 73.5 | $20.00 | |
| 10 | DeepSeek V4.1 Flash 1017DeepSeek | 1358 | 170 | 70.3 | $0.60 | |
| 11 | Muse Spark 1.3Meta | 1358 | 5,508 | 57.4 | $4.25 | |
| 12 | DeepSeek V4.1 FlashDeepSeek | 1355 | 5,318 | 62.5 | $0.84 | |
| 13 | Claude Haiku 5.5Anthropic | 1342 | 771 | 70.7 | $0.50 | |
| 14 | Qwen3.8 Max (0902)Qwen | 1336 | 3,294 | 62.4 | $6.00 | |
| 15 | GLM 5.1Z.ai | 1336 | Not published | 49.2 | $4.40 | |
| 16 | Claude Opus 5Anthropic | 1334 | 3,734 | 83.0 | $25.00 | |
| 17 | GLM 5.3Z.ai | 1332 | Not published | 70.6 | $4.40 | |
| 18 | MiMo-V2.6-ProXiaomi | 1330 | Not published | 65.4 | $0.87 | |
| 19 | Claude Fable 5Anthropic | 1318 | 4,617 | 79.3 | $50.00 | |
| 20 | GLM 5.3 FlashZ.ai | 1316 | Not published | 57.7 | $0.50 | |
| 21 | GPT-6 SolOpenAI | 1314 | 499 | 76.0 | $10.00 | |
| 22 | Gemini 3.7 FlashGoogle | 1307 | 7,637 | 58.0 | $3.75 | |
| 23 | Muse Spark 1.2Meta | 1301 | 6,599 | 55.7 | $4.25 | |
| 24 | Gemini 3.8 FlashGoogle | 1294 | 6,304 | 61.6 | $3.75 | |
| 25 | GLM 5.2Z.ai | 1287 | 7,912 | 61.1 | $4.40 | |
| 26 | Grok 4.6xAI | 1277 | Not published | 65.7 | $6.00 | |
| 27 | Grok 4.7xAI | 1274 | 332 | 66.7 | $6.00 | |
| 28 | Claude Opus 4.6Anthropic | 1273 | 5,862 | 57.2 | $25.00 | |
| 29 | Gemini 3.6 FlashGoogle | 1271 | 7,934 | 62.5 | $3.75 | |
| 30 | Kimi K2.6Moonshot AI | 1267 | 13,980 | 52.1 | $4.00 | |
| 31 | Qwen3.7 MaxQwen | 1265 | 6,908 | 54.1 | $4.425 | |
| 32 | Claude Sonnet 5Anthropic | 1261 | 5,329 | 57.7 | $10.00 | |
| 33 | Muse Spark 1.1Meta | 1260 | Not published | 70.3 | $4.25 | |
| 34 | Grok 4.5xAI | 1257 | Not published | 58.6 | $6.00 | |
| 35 | DeepSeek V4 Pro 0423DeepSeek | 1253 | 4,513 | 38.7 | $2.976 | |
| 36 | Mistral Large 4Mistral AI | 1252 | 1,344 | 56.4 | $4.18 | |
| 37 | GLM 5 TurboZ.ai | 1247 | 5,658 | Not benchmarked | $4.00 | |
| 38 | Gemini 3.5 FlashGoogle | 1243 | 7,373 | 39.8 | $9.00 | |
| 39 | MiMo-V2.5-ProXiaomi | 1243 | 4,252 | 31.6 | $0.9152 | |
| 40 | Kimi K2.7 CodeMoonshot AI | 1240 | 6,668 | 53.4 | $4.00 | |
| 41 | Qwen3.7 PlusQwen | 1238 | 5,114 | 33.2 | $1.28 | |
| 42 | Claude Sonnet 4.6Anthropic | 1234 | 5,715 | 57.3 | $15.00 | |
| 43 | Gemini 3.1 Pro PreviewGoogle | 1234 | 9,108 | 32.3 | $12.00 | |
| 44 | Hy3Tencent | 1232 | 4,176 | Not benchmarked | $0.528 | |
| 45 | GLM 5Z.ai | 1228 | 9,258 | 41.7 | $3.20 | |
| 46 | Claude Opus 4.8Anthropic | 1222 | 7,351 | 65.0 | $25.00 | |
| 47 | GPT-6 LunaOpenAI | 1218 | 711 | 57.3 | $0.50 | |
| 48 | MiMo-V2.5Xiaomi | 1215 | 11,017 | 50.0 | $0.336 | |
| 49 | Claude Opus 4.5Anthropic | 1213 | 6,015 | 44.0 | $25.00 | |
| 50 | DeepSeek V4 Flash 0731DeepSeek | 1208 | 1,232 | 51.8 | $0.396 | |
| 51 | MiniMax M3MiniMax | 1208 | 4,898 | 39.2 | $1.20 | |
| 52 | GLM 5V TurboZ.ai | 1208 | 13,392 | Not benchmarked | $4.00 | |
| 53 | Kimi K2.5Moonshot AI | 1204 | 11,631 | 38.6 | $2.85 | |
| 54 | Qwen3.6 PlusQwen | 1202 | 12,707 | 44.1 | $1.95 | |
| 55 | GPT-5.5OpenAI | 1197 | 5,749 | 71.3 | $30.00 | |
| 56 | MiniMax M2.7MiniMax | 1195 | Not published | 36.2 | $1.20 | |
| 57 | Grok 4.20xAI | 1193 | Not published | 33.3 | $2.50 | |
| 58 | GLM 4.7Z.ai | 1190 | 7,009 | 55.1 | $2.20 | |
| 59 | DeepSeek V4 Flash 0423DeepSeek | 1188 | 8,028 | Not benchmarked | $0.28 | |
| 60 | Gemini 3 Flash PreviewGoogle | 1183 | 467 | 43.7 | $3.00 | |
| 61 | Solar Pro 4Upstage | 1178 | 2,694 | Not benchmarked | $0.60 | |
| 62 | MiniMax M2.5MiniMax | 1170 | Not published | 37.2 | $1.20 | |
| 63 | GLM 4.5Z.ai | 1170 | 1,878 | Not benchmarked | $2.20 | |
| 64 | MiniMax M2.1MiniMax | 1164 | 2,159 | Not benchmarked | $1.20 | |
| 65 | Qwen3.5 397B A17BQwen | 1160 | 960 | Not benchmarked | $2.34 | |
| 66 | T InklingThinking Machines | 1160 | 2,334 | 31.5 | $4.05 | |
| 67 | Claude Sonnet 4.5Anthropic | 1157 | 7,375 | 39.5 | $15.00 | |
| 68 | Claude Opus 4.1Anthropic | 1152 | 5,947 | Not benchmarked | $75.00 | |
| 69 | DeepSeek V3.2 ExpDeepSeek | 1148 | 1,976 | Not benchmarked | $0.41 | |
| 70 | DeepSeek V3.1 TerminusDeepSeek | 1141 | Not published | Not benchmarked | $1.0142 | |
| 71 | Claude Opus 4RetiredAnthropic | 1138 | 1,469 | Not benchmarked | $75.00 | |
| 72 | GPT-5.6 TerraOpenAI | 1134 | 17,454 | 63.9 | $12.00 | |
| 73 | DeepSeek V3.2DeepSeek | 1132 | 4,161 | 30.0 | $0.88 | |
| 74 | Nemotron 3 UltraNVIDIA | 1131 | 4,098 | 30.0 | $2.40 | |
| 75 | Step 3.7 FlashStepFun | 1129 | 6,914 | Not benchmarked | $1.15 | |
| 76 | GLM 4.6Z.ai | 1128 | 1,671 | 8.8 | $2.20 | |
| 77 | Nemotron 3.5 Super VLNVIDIA | 1127 | 212 | Not benchmarked | $0.50 | |
| 78 | Grok 4.3xAI | 1124 | 11,709 | 37.4 | $2.50 | |
| 79 | GLM 4.5 AirZ.ai | 1122 | 1,475 | Not benchmarked | $1.10 | |
| 80 | GPT-5.6 LunaOpenAI | 1121 | 3,784 | 58.0 | $1.20 | |
| 81 | GLM 4.7 FlashZ.ai | 1120 | 1,460 | Not benchmarked | $0.40 | |
| 82 | Qwen3.5 Plus 2026-02-15Qwen | 1118 | 2,650 | Not benchmarked | $1.56 | |
| 83 | R1 0528DeepSeek | 1110 | 1,630 | Not benchmarked | $2.18 | |
| 84 | Mistral Large 3 2512Mistral AI | 1099 | 4,549 | Not benchmarked | $1.50 | |
| 85 | MiniMax M2MiniMax | 1098 | Not published | Not benchmarked | $1.02 | |
| 86 | Claude 3.7 SonnetRetiredAnthropic | 1093 | 1,371 | Not benchmarked | $15.00 | |
| 87 | DeepSeek V3DeepSeek | 1086 | 1,821 | Not benchmarked | $0.9594 | |
| 88 | Mistral Medium 3Mistral AI | 1083 | 512 | Not benchmarked | $2.00 | |
| 89 | Trinity Large ThinkingArcee AI | 1081 | 2,117 | Not benchmarked | $0.80 | |
| 90 | Mistral Medium 3.1Mistral AI | 1081 | 4,180 | Not benchmarked | $2.00 | |
| 91 | Gemini 2.5 ProGoogle | 1080 | 644 | 19.1 | $10.00 | |
| 92 | GPT-5.2OpenAI | 1079 | Not published | 55.9 | $14.00 | |
| 93 | DeepSeek V3.1DeepSeek | 1077 | Not published | Not benchmarked | $1.50 | |
| 94 | Claude Haiku 4.5Anthropic | 1071 | 6,345 | 35.6 | $5.00 | |
| 95 | GPT-5.4OpenAI | 1065 | 4,300 | 64.4 | $15.00 | |
| 96 | Mistral Medium 3.5Mistral AI | 1060 | 2,065 | 25.5 | $7.50 | |
| 97 | GPT-5.1OpenAI | 1060 | 1,850 | 40.4 | $10.00 | |
| 98 | GPT-5OpenAI | 1055 | 3,973 | 38.8 | $10.00 | |
| 99 | Grok 3RetiredxAI | 1049 | 2,656 | Not benchmarked | $15.00 | |
| 100 | Gemini 3.1 Flash Lite PreviewGoogle | 1047 | 3,839 | 22.7 | $1.50 | |
| 101 | GPT-5 MiniOpenAI | 1036 | 5,580 | 34.2 | $2.00 | |
| 102 | Ministral 3 8B 2512Mistral AI | 1031 | 238 | Not benchmarked | $0.15 | |
| 103 | Gemini 2.5 FlashGoogle | 1020 | 495 | Not benchmarked | $2.50 | |
| 104 | Codestral 2508Mistral AI | 1019 | 445 | Not benchmarked | $0.90 | |
| 105 | GPT-5.3-CodexOpenAI | 1007 | 2,219 | 60.0 | $14.00 | |
| 106 | Qwen3 235B A22B Thinking 2507Qwen | 997 | 384 | Not benchmarked | $3.50 | |
| 107 | Qwen3 235B A22B Instruct 2507Qwen | 994 | 452 | Not benchmarked | $0.75 | |
| 108 | Ministral 3 14B 2512Mistral AI | 989 | 212 | Not benchmarked | $0.20 | |
| 109 | GPT-5.1-Codex-MiniOpenAI | 988 | 4,527 | Not benchmarked | $2.00 | |
| 110 | Mercury 2Inception | 982 | 3,792 | Not benchmarked | $0.75 | |
| 111 | GPT-5 NanoOpenAI | 965 | 383 | Not benchmarked | $0.40 | |
| 112 | Ministral 3 3B 2512Mistral AI | 962 | 195 | Not benchmarked | $0.10 | |
| 113 | GPT-4.1 NanoOpenAI | 924 | 213 | Not benchmarked | $0.40 | |
| 114 | gpt-oss-120bOpenAI | 901 | 313 | Not benchmarked | $0.50 | |
| 115 | Llama 4 MaverickMeta Llama | 900 | 164 | Not benchmarked | $1.00 | |
| 116 | GPT-4oOpenAI | 868 | 194 | Not benchmarked | $10.00 | |
| 117 | o4 MiniOpenAI | 854 | Not published | Not benchmarked | $4.40 | |
| 118 | GPT-4.1OpenAI | 849 | 181 | Not benchmarked | $8.00 | |
| 119 | GPT-4.1 MiniOpenAI | 836 | 151 | Not benchmarked | $1.60 |
Only models in our catalog are listed, and each appears once with its best published run. 145 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.