Design Arena: UI components leaderboard
- GPT-6 Astra leads
- 1393
- models measured
- 121
- lead over #2
- +8 Elo
- median
- 1210
Leader on Design Arena: UI components · read October 22
On October 22, 2026, 124 of our catalog's models had a Design Arena: UI components result, and OpenAI GPT-6 Astra led with 1393, ahead of Anthropic Claude Opus 5.5 at 1385.
Design Arena shows people two designs made by two models from the same prompt (a UI component) 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: UI components 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 | 1393 | 586 | 85.1 | $50.00 | |
| 2 | Claude Opus 5.5Anthropic | 1385 | 394 | 88.8 | $20.00 | |
| 3 | Muse WonderMeta | 1379 | 290 | 71.9 | $20.00 | |
| 4 | Claude Sonnet 5.5Anthropic | 1368 | 370 | 87.3 | $10.00 | |
| 5 | Muse Spark 1.3Meta | 1349 | 1,370 | 57.4 | $4.25 | |
| 6 | Claude Opus 5Anthropic | 1346 | 1,622 | 83.0 | $25.00 | |
| 7 | Kimi K3Moonshot AI | 1341 | 501 | 62.3 | $15.00 | |
| 8 | MiMo-V2.6-ProXiaomi | 1340 | Not published | 65.4 | $0.87 | |
| 9 | Qwen3.8 Max (0902)Qwen | 1338 | 1,003 | 62.4 | $6.00 | |
| 10 | Claude Fable 5.1Anthropic | 1336 | 684 | 84.2 | $50.00 | |
| 11 | GPT-5.6 SolOpenAI | 1335 | 831 | 73.5 | $20.00 | |
| 12 | Gemini 4 ArgonGoogle | 1333 | 644 | 81.0 | $10.00 | |
| 13 | GPT-6.1 SolOpenAI | 1317 | 211 | 82.4 | $10.00 | |
| 14 | Claude Fable 5Anthropic | 1316 | 1,547 | 79.3 | $50.00 | |
| 15 | Gemini 3.8 FlashGoogle | 1316 | 1,778 | 61.6 | $3.75 | |
| 16 | GLM 5.3 FlashZ.ai | 1316 | Not published | 57.7 | $0.50 | |
| 17 | GLM 5.3Z.ai | 1314 | Not published | 70.6 | $4.40 | |
| 18 | DeepSeek V4.1 FlashDeepSeek | 1312 | 1,205 | 62.5 | $0.84 | |
| 19 | DeepSeek V4.1 Flash 1017DeepSeek | 1303 | 163 | 70.3 | $0.60 | |
| 20 | Muse Spark 1.2Meta | 1298 | 1,922 | 55.7 | $4.25 | |
| 21 | Gemini 3.7 FlashGoogle | 1296 | 1,926 | 58.0 | $3.75 | |
| 22 | GPT-6 LunaOpenAI | 1295 | 1,456 | 57.3 | $0.50 | |
| 23 | Gemini 3.6 FlashGoogle | 1294 | 1,893 | 62.5 | $3.75 | |
| 24 | Grok 4.6xAI | 1294 | Not published | 65.7 | $6.00 | |
| 25 | GLM 5.2Z.ai | 1294 | 1,848 | 61.1 | $4.40 | |
| 26 | GLM 5.1Z.ai | 1292 | Not published | 49.2 | $4.40 | |
| 27 | Claude Haiku 5.5Anthropic | 1291 | 309 | 70.7 | $0.50 | |
| 28 | GPT-6 SolOpenAI | 1290 | 915 | 76.0 | $10.00 | |
| 29 | Muse Spark 1.1Meta | 1289 | Not published | 70.3 | $4.25 | |
| 30 | Claude Opus 4.6Anthropic | 1281 | 2,319 | 57.2 | $25.00 | |
| 31 | Grok 4.5xAI | 1280 | Not published | 58.6 | $6.00 | |
| 32 | Claude Sonnet 5Anthropic | 1277 | 2,573 | 57.7 | $10.00 | |
| 33 | Grok 4.7xAI | 1275 | 311 | 66.7 | $6.00 | |
| 34 | Mistral Large 4Mistral AI | 1272 | 416 | 56.4 | $4.18 | |
| 35 | Claude Sonnet 4.6Anthropic | 1269 | 2,167 | 57.3 | $15.00 | |
| 36 | Gemini 3.5 FlashGoogle | 1269 | 3,230 | 39.8 | $9.00 | |
| 37 | Qwen3.7 MaxQwen | 1267 | 2,147 | 54.1 | $4.425 | |
| 38 | Kimi K2.6Moonshot AI | 1263 | 3,160 | 52.1 | $4.00 | |
| 39 | GLM 5 TurboZ.ai | 1260 | 2,930 | Not benchmarked | $4.00 | |
| 40 | Qwen3.7 PlusQwen | 1258 | 1,391 | 33.2 | $1.28 | |
| 41 | GPT-5.5OpenAI | 1257 | 2,533 | 71.3 | $30.00 | |
| 42 | GPT-5.6 TerraOpenAI | 1256 | 7,007 | 63.9 | $12.00 | |
| 43 | Claude Opus 4.8Anthropic | 1255 | 2,361 | 65.0 | $25.00 | |
| 44 | Mercury 3Inception | 1255 | 296 | Not benchmarked | $0.35 | |
| 45 | Kimi K2.7 CodeMoonshot AI | 1255 | 1,967 | 53.4 | $4.00 | |
| 46 | MiMo-V2.5Xiaomi | 1255 | 4,290 | 50.0 | $0.336 | |
| 47 | GPT-5.6 LunaOpenAI | 1248 | 1,831 | 58.0 | $1.20 | |
| 48 | MiMo-V2.5-ProXiaomi | 1246 | 1,525 | 31.6 | $0.9152 | |
| 49 | MiniMax M3MiniMax | 1242 | 1,865 | 39.2 | $1.20 | |
| 50 | Claude Opus 4.5Anthropic | 1235 | 2,085 | 44.0 | $25.00 | |
| 51 | Kimi K2.5Moonshot AI | 1234 | 3,531 | 38.6 | $2.85 | |
| 52 | GPT-5.1-CodexOpenAI | 1234 | 115 | 15.3 | $10.00 | |
| 53 | DeepSeek V4 Flash 0731DeepSeek | 1233 | 496 | 51.8 | $0.396 | |
| 54 | Qwen3.6 PlusQwen | 1230 | 4,387 | 44.1 | $1.95 | |
| 55 | GLM 5Z.ai | 1227 | 3,908 | 41.7 | $3.20 | |
| 56 | Gemini 3.1 Pro PreviewGoogle | 1224 | 3,328 | 32.3 | $12.00 | |
| 57 | MiniMax M2.1MiniMax | 1220 | 746 | Not benchmarked | $1.20 | |
| 58 | DeepSeek V4 Pro 0423DeepSeek | 1219 | 2,145 | 38.7 | $2.976 | |
| 59 | GPT-5.4OpenAI | 1212 | 129 | 64.4 | $15.00 | |
| 60 | MiniMax M2.7MiniMax | 1210 | Not published | 36.2 | $1.20 | |
| 61 | GLM 5V TurboZ.ai | 1210 | 3,245 | Not benchmarked | $4.00 | |
| 62 | GLM 4.7 FlashZ.ai | 1205 | 583 | Not benchmarked | $0.40 | |
| 63 | Hy3Tencent | 1204 | 1,572 | Not benchmarked | $0.528 | |
| 64 | Grok 4.20xAI | 1200 | Not published | 33.3 | $2.50 | |
| 65 | T InklingThinking Machines | 1198 | 1,434 | 31.5 | $4.05 | |
| 66 | GLM 4.7Z.ai | 1196 | 2,898 | 55.1 | $2.20 | |
| 67 | GPT-5.2OpenAI | 1190 | Not published | 55.9 | $14.00 | |
| 68 | Grok 4.3xAI | 1190 | 3,972 | 37.4 | $2.50 | |
| 69 | DeepSeek V3.1 TerminusDeepSeek | 1182 | Not published | Not benchmarked | $1.0142 | |
| 70 | GPT-5OpenAI | 1177 | 458 | 38.8 | $10.00 | |
| 71 | Step 3.7 FlashStepFun | 1177 | 2,168 | Not benchmarked | $1.15 | |
| 72 | Qwen3.5 Plus 2026-02-15Qwen | 1175 | 1,238 | Not benchmarked | $1.56 | |
| 73 | MiniMax M2.5MiniMax | 1173 | Not published | 37.2 | $1.20 | |
| 74 | Claude Sonnet 4.5Anthropic | 1170 | 2,412 | 39.5 | $15.00 | |
| 75 | DeepSeek V3.2 ExpDeepSeek | 1164 | 752 | Not benchmarked | $0.41 | |
| 76 | GPT-5.1OpenAI | 1163 | 854 | 40.4 | $10.00 | |
| 77 | Claude Opus 4.1Anthropic | 1162 | 431 | Not benchmarked | $75.00 | |
| 78 | DeepSeek V4 Flash 0423DeepSeek | 1162 | 1,577 | Not benchmarked | $0.28 | |
| 79 | Qwen3.5 397B A17BQwen | 1160 | 488 | Not benchmarked | $2.34 | |
| 80 | Nemotron 3.5 Super VLNVIDIA | 1153 | 130 | Not benchmarked | $0.50 | |
| 81 | Claude Opus 4RetiredAnthropic | 1150 | 512 | Not benchmarked | $75.00 | |
| 82 | Claude 3.7 SonnetRetiredAnthropic | 1148 | 535 | Not benchmarked | $15.00 | |
| 83 | Solar Pro 4Upstage | 1147 | 1,049 | Not benchmarked | $0.60 | |
| 84 | DeepSeek V3.2DeepSeek | 1146 | 1,615 | 30.0 | $0.88 | |
| 85 | GLM 4.6Z.ai | 1145 | 686 | 8.8 | $2.20 | |
| 86 | GLM 4.5Z.ai | 1141 | 607 | Not benchmarked | $2.20 | |
| 87 | Gemini 2.5 ProGoogle | 1137 | 572 | 19.1 | $10.00 | |
| 88 | MiniMax M2MiniMax | 1137 | Not published | Not benchmarked | $1.02 | |
| 89 | Nemotron 3 UltraNVIDIA | 1136 | 1,542 | 30.0 | $2.40 | |
| 90 | GPT-5.3-CodexOpenAI | 1134 | 1,051 | 60.0 | $14.00 | |
| 91 | GLM 4.5 AirZ.ai | 1120 | 465 | Not benchmarked | $1.10 | |
| 92 | Qwen3 Coder 480B A35BQwen | 1110 | 207 | Not benchmarked | $1.50 | |
| 93 | GPT-5 MiniOpenAI | 1103 | 2,100 | 34.2 | $2.00 | |
| 94 | R1 0528DeepSeek | 1100 | 530 | Not benchmarked | $2.18 | |
| 95 | Mistral Large 3 2512Mistral AI | 1099 | 1,735 | Not benchmarked | $1.50 | |
| 96 | Claude Haiku 4.5Anthropic | 1097 | 1,865 | 35.6 | $5.00 | |
| 97 | Mistral Medium 3.1Mistral AI | 1097 | 1,708 | Not benchmarked | $2.00 | |
| 98 | DeepSeek V3DeepSeek | 1090 | 581 | Not benchmarked | $0.9594 | |
| 99 | DeepSeek V3.1DeepSeek | 1085 | Not published | Not benchmarked | $1.50 | |
| 100 | GPT-5.1-Codex-MiniOpenAI | 1078 | 1,837 | Not benchmarked | $2.00 | |
| 101 | Mistral Medium 3.5Mistral AI | 1076 | 889 | 25.5 | $7.50 | |
| 102 | Gemini 3.1 Flash Lite PreviewGoogle | 1069 | 1,521 | 22.7 | $1.50 | |
| 103 | GPT-5 NanoOpenAI | 1064 | 268 | Not benchmarked | $0.40 | |
| 104 | Qwen3 Coder 30B A3B InstructQwen | 1043 | 122 | Not benchmarked | $0.44 | |
| 105 | Trinity Large ThinkingArcee AI | 1042 | 841 | Not benchmarked | $0.80 | |
| 106 | Grok 3RetiredxAI | 1034 | 1,113 | Not benchmarked | $15.00 | |
| 107 | Kimi K2 0711Moonshot AI | 1027 | 138 | Not benchmarked | $2.30 | |
| 108 | Mistral Medium 3Mistral AI | 1022 | 365 | Not benchmarked | $2.00 | |
| 109 | o3OpenAI | 1017 | Not published | Not benchmarked | $8.00 | |
| 110 | Codestral 2508Mistral AI | 1014 | 257 | Not benchmarked | $0.90 | |
| 111 | Gemini 2.5 FlashGoogle | 1002 | 298 | Not benchmarked | $2.50 | |
| 112 | GPT-4.1OpenAI | 997 | 155 | Not benchmarked | $8.00 | |
| 113 | Mercury 2Inception | 996 | 1,549 | Not benchmarked | $0.75 | |
| 114 | o4 MiniOpenAI | 983 | Not published | Not benchmarked | $4.40 | |
| 115 | Qwen3 235B A22B Instruct 2507Qwen | 962 | 380 | Not benchmarked | $0.75 | |
| 116 | GPT-4.1 MiniOpenAI | 958 | 172 | Not benchmarked | $1.60 | |
| 117 | Qwen3 30B A3BQwen | 943 | 264 | Not benchmarked | $0.50 | |
| 118 | Qwen3 235B A22B Thinking 2507Qwen | 942 | 244 | Not benchmarked | $3.50 | |
| 119 | gpt-oss-120bOpenAI | 924 | 269 | Not benchmarked | $0.50 | |
| 120 | GPT-4.1 NanoOpenAI | 913 | 173 | Not benchmarked | $0.40 | |
| 121 | Mistral Small 3.2 24BMistral AI | 907 | Not published | Not benchmarked | $0.30 | |
| 122 | Llama 4 MaverickMeta Llama | 896 | 184 | Not benchmarked | $1.00 | |
| 123 | GPT-4oOpenAI | 882 | 197 | Not benchmarked | $10.00 | |
| 124 | Llama 4 ScoutMeta Llama | 765 | 157 | Not benchmarked | $0.50 |
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.