Design Arena: ASCII art leaderboard
- GPT-6 Astra leads
- 1359
- models measured
- 74
- lead over #2
- +7 Elo
- median
- 1166
Leader on Design Arena: ASCII art · read October 21
On October 21, 2026, 74 of our catalog's models had a Design Arena: ASCII art result, and OpenAI GPT-6 Astra led with 1359.
Design Arena shows people two designs made by two models from the same prompt (a piece of ASCII art) and asks which is better. The number is the Elo rating built from those votes, as Design Arena publishes it.
Role in the index
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Battles | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 1359 | 170 | 88.7 | $50.00 | |
| 2 | Claude Opus 5Anthropic | 1352 | 227 | 81.8 | $25.00 | |
| 3 | Claude Fable 5Anthropic | 1328 | 298 | 82.1 | $50.00 | |
| 4 | Claude Fable 5.1Anthropic | 1324 | 137 | 89.1 | $50.00 | |
| 5 | GPT-5.6 SolOpenAI | 1302 | 248 | 76.7 | $20.00 | |
| 6 | Muse Spark 1.2Meta | 1284 | 553 | 65.1 | $4.25 | |
| 7 | Gemini 3.6 FlashGoogle | 1281 | 354 | 48.3 | $3.75 | |
| 8 | Muse Spark 1.3Meta | 1280 | 216 | 79.7 | $4.25 | |
| 9 | Grok 4.6xAI | 1278 | Not published | 76.4 | $6.00 | |
| 10 | Claude Opus 4.8Anthropic | 1276 | 448 | 66.0 | $25.00 | |
| 11 | Gemini 3.7 FlashGoogle | 1276 | 545 | 69.9 | $3.75 | |
| 12 | Gemini 3.1 Pro PreviewGoogle | 1270 | 1,359 | 35.3 | $12.00 | |
| 13 | Muse Spark 1.1Meta | 1270 | Not published | 48.9 | $4.25 | |
| 14 | GLM 5.3 FlashZ.ai | 1269 | Not published | 71.0 | $0.50 | |
| 15 | DeepSeek V4.1 FlashDeepSeek | 1265 | 288 | 77.2 | $0.525 | |
| 16 | Gemini 3.8 FlashGoogle | 1264 | 269 | 70.5 | $4.125 | |
| 17 | Grok 4.5xAI | 1257 | Not published | 75.5 | $6.00 | |
| 18 | Claude Opus 4.6Anthropic | 1254 | 412 | 66.9 | $25.00 | |
| 19 | GPT-5.5OpenAI | 1254 | 555 | 67.8 | $30.00 | |
| 20 | Gemini 3.5 FlashGoogle | 1253 | 1,387 | 42.8 | $9.00 | |
| 21 | DeepSeek V4.1 Flash 1017DeepSeek | 1244 | 100 | 83.3 | $0.60 | |
| 22 | Claude Sonnet 4.6Anthropic | 1235 | 411 | 57.8 | $15.00 | |
| 23 | GPT-6 SolOpenAI | 1226 | 241 | 82.8 | $10.00 | |
| 24 | GLM 5.3Z.ai | 1213 | Not published | 73.1 | $1.40 | |
| 25 | Qwen3.7 MaxQwen | 1211 | 788 | 45.1 | $4.425 | |
| 26 | Claude Sonnet 5Anthropic | 1207 | 707 | 63.5 | $10.00 | |
| 27 | Claude Sonnet 4.5Anthropic | 1204 | 531 | 59.1 | $15.00 | |
| 28 | Mistral Large 4Mistral AI | 1203 | 161 | 55.3 | $4.18 | |
| 29 | Qwen3.8 Max (0902)Qwen | 1201 | 248 | 63.5 | $6.00 | |
| 30 | Kimi K2.7 CodeMoonshot AI | 1196 | 505 | 39.1 | $3.35 | |
| 31 | Claude Opus 4.5Anthropic | 1190 | 429 | 66.8 | $25.00 | |
| 32 | GLM 5.2Z.ai | 1190 | 583 | 57.7 | $7.00 | |
| 33 | GPT-5.4OpenAI | 1188 | 1,312 | 37.7 | $15.00 | |
| 34 | GPT-5.6 TerraOpenAI | 1177 | 1,415 | 69.0 | $12.00 | |
| 35 | Grok 4.20xAI | 1175 | Not published | Not benchmarked | $2.50 | |
| 36 | Claude Opus 4.1Anthropic | 1167 | 458 | 27.2 | $75.00 | |
| 37 | Gemini 3.1 Flash Lite PreviewGoogle | 1166 | 1,211 | 8.1 | $1.50 | |
| 38 | Kimi K2.5Moonshot AI | 1165 | 931 | 58.5 | $2.50 | |
| 39 | GLM 4.7Z.ai | 1165 | 1,137 | 33.6 | $1.98 | |
| 40 | Kimi K2.6Moonshot AI | 1164 | 120 | 44.2 | $3.41 | |
| 41 | MiniMax M3MiniMax | 1161 | 503 | 57.8 | $1.20 | |
| 42 | GLM 5 TurboZ.ai | 1159 | 395 | Not benchmarked | $4.00 | |
| 43 | Hy3Tencent | 1157 | 201 | 43.9 | $0.528 | |
| 44 | MiMo-V2.5-ProXiaomi | 1157 | 262 | 40.0 | $0.609 | |
| 45 | GPT-6 LunaOpenAI | 1156 | 278 | 71.4 | $0.50 | |
| 46 | GPT-5.2OpenAI | 1153 | Not published | 58.9 | $14.00 | |
| 47 | Step 3.7 FlashStepFun | 1152 | 609 | Not benchmarked | $1.15 | |
| 48 | GPT-5.3-CodexOpenAI | 1151 | 258 | 29.5 | $14.00 | |
| 49 | DeepSeek V4 Pro 0423DeepSeek | 1148 | 679 | 35.3 | $1.9105 | |
| 50 | GLM 5Z.ai | 1146 | 984 | 59.7 | $3.20 | |
| 51 | GPT-5.6 LunaOpenAI | 1144 | 316 | 61.5 | $1.20 | |
| 52 | Qwen3.7 PlusQwen | 1144 | 254 | Not benchmarked | $1.28 | |
| 53 | GLM 5.1Z.ai | 1142 | Not published | 44.0 | $4.40 | |
| 54 | MiMo-V2.5Xiaomi | 1141 | 361 | 34.0 | $0.238 | |
| 55 | GPT-5OpenAI | 1140 | 312 | 72.2 | $10.00 | |
| 56 | Claude Haiku 4.5Anthropic | 1139 | 353 | 49.8 | $5.00 | |
| 57 | MiniMax M2.7MiniMax | 1136 | Not published | 28.4 | $1.20 | |
| 58 | Grok 4.3xAI | 1125 | 603 | 22.5 | $2.50 | |
| 59 | GPT-5 MiniOpenAI | 1124 | 417 | 59.8 | $2.00 | |
| 60 | GPT-5.1OpenAI | 1118 | 140 | 53.3 | $10.00 | |
| 61 | GLM 5V TurboZ.ai | 1118 | 479 | 28.8 | $4.00 | |
| 62 | Qwen3.6 PlusQwen | 1115 | 458 | 37.4 | $1.95 | |
| 63 | DeepSeek V4 Flash 0423DeepSeek | 1109 | 300 | 54.6 | $0.182 | |
| 64 | GPT-5.1-Codex-MiniOpenAI | 1104 | 309 | 6.2 | $2.00 | |
| 65 | Nemotron 3 UltraNVIDIA | 1101 | 293 | Not benchmarked | $2.40 | |
| 66 | Qwen3.5 Plus 2026-02-15Qwen | 1093 | 296 | Not benchmarked | $1.56 | |
| 67 | DeepSeek V4 Flash 0731DeepSeek | 1090 | 105 | Not benchmarked | $1.32 | |
| 68 | T InklingThinking Machines | 1089 | 220 | 49.7 | $4.05 | |
| 69 | DeepSeek V3.2DeepSeek | 1082 | 259 | 54.4 | $0.42 | |
| 70 | Mistral Large 3 2512Mistral AI | 1068 | 340 | Not benchmarked | $1.50 | |
| 71 | Mistral Medium 3.5Mistral AI | 1062 | 205 | 9.2 | $7.50 | |
| 72 | Trinity Large ThinkingArcee AI | 1041 | 116 | 5.4 | $0.80 | |
| 73 | Mistral Medium 3.1Mistral AI | 997 | 357 | Not benchmarked | $2.00 | |
| 74 | Mercury 2Inception | 990 | 213 | 0.0 | $0.75 |
Only models in our catalog are listed, and each appears once with its best published run. 141 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.