---
title: "Best AI for design"
description: "Models ranked on the LMArena webdev arenas, covering React, HTML and screenshot to interface, with coverage shown for every row."
url: "https://aldena.ai/best-ai-for-design"
---

# Best AI for design

Design here means producing an interface a person prefers, not writing about design.

Aldena runs these models inside your team rooms. [See what each one costs](https://aldena.ai/models).

| rank | model | vendor | score | coverage (benchmarks) | price (in / out) | LMArena webdev (elo) | LMArena webdev React (elo) | LMArena webdev HTML (elo) | LMArena image to webdev (elo) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Claude Opus 5 (max) | Anthropic | 82.4 | 4 of 4 | $5.00 / $25.00 | 1692 | 1711 | 1679 | 1670 |
| 2 | Qwen3.8 Max | Qwen | 80.5 | 4 of 4 | $2.00 / $6.00 | 1667 | 1675 | 1627 | 1631 |
| 3 | Claude Fable 5 | Anthropic | 79.8 | 4 of 4 | $10.00 / $50.00 | 1627 | 1630 | 1669 | 1626 |
| 4 | Kimi K3 (max) | MoonshotAI | 79.2 | 4 of 4 | $3.00 / $15.00 | 1674 | 1692 | 1648 | 1570 |
| 5 | GPT-5.6 Sol (xhigh) | OpenAI | 78.4 | 4 of 4 | $5.00 / $30.00 | 1622 | 1628 | 1622 | 1581 |
| 6 | Claude Opus 5 (high) | Anthropic | 77.6 | 3 of 4 | $5.00 / $25.00 | 1663 | 1679 | 1653 | — |
| 7 | Grok 4.6 (high) | SpaceXAI | 76.6 | 3 of 4 | $2.00 / $6.00 | 1631 | 1640 | 1635 | — |
| 8 | Grok 4.5 | SpaceXAI | 75.1 | 4 of 4 | $2.00 / $6.00 | 1555 | 1556 | 1576 | 1580 |
| 9 | Gemini 3.7 Flash (high) | Google | 74.3 | 3 of 4 | $0.375 / $1.875 | 1587 | 1582 | 1621 | — |
| 10 | DeepSeek V4 Pro (high) | DeepSeek | 74.2 | 3 of 4 | $1.168 / $2.336 | 1584 | 1594 | 1577 | — |
| 11 | Claude Opus 4.7 | Anthropic | 73.8 | 4 of 4 | $5.00 / $25.00 | 1558 | 1555 | 1561 | 1565 |
| 12 | GLM 5.2 (max) | Z.ai | 72.6 | 3 of 4 | $0.462 / $1.452 | 1585 | 1596 | 1538 | — |
| 13 | Claude Opus 4.8 (high) | Anthropic | 72.5 | 3 of 4 | $5.00 / $25.00 | 1564 | 1563 | 1558 | — |
| 14 | Claude Opus 4.7 (high) | Anthropic | 71.7 | 3 of 4 | $5.00 / $25.00 | 1557 | 1558 | 1553 | — |
| 15 | DeepSeek V4 Flash 0423 (high) | DeepSeek | 71.6 | 3 of 4 | $0.0643 / $0.1285 | 1581 | 1588 | 1538 | — |
| 16 | Claude Opus 4.6 (high) | Anthropic | 70.7 | 3 of 4 | $5.00 / $25.00 | 1545 | 1538 | 1559 | — |
| 17 | Muse Spark 1.1 | Meta | 70.2 | 4 of 4 | $1.25 / $4.25 | 1538 | 1529 | 1540 | 1544 |
| 18 | Claude Opus 4.6 | Anthropic | 69.5 | 4 of 4 | $5.00 / $25.00 | 1537 | 1529 | 1550 | 1538 |
| 19 | Claude Sonnet 5 (high) | Anthropic | 69.3 | 4 of 4 | $2.00 / $10.00 | 1540 | 1545 | 1520 | 1540 |
| 20 | Claude Opus 4.8 | Anthropic | 68.0 | 3 of 4 | $5.00 / $25.00 | 1539 | 1542 | 1525 | — |
| 21 | Muse Spark 1.2 (xhigh) | Meta | 67.3 | 3 of 4 | $1.25 / $4.25 | 1535 | 1530 | 1537 | — |
| 22 | Gemini 3.6 Flash (high) | Google | 66.5 | 3 of 4 | $0.75 / $3.75 | 1537 | 1529 | 1535 | — |
| 23 | Claude Sonnet 4.6 | Anthropic | 66.2 | 4 of 4 | $3.00 / $15.00 | 1524 | 1513 | 1528 | 1530 |
| 24 | GPT-5.6 Terra (xhigh) | OpenAI | 65.6 | 4 of 4 | $1.00 / $6.00 | 1521 | 1512 | 1541 | 1525 |
| 25 | GPT-5.5 (xhigh) | OpenAI | 64.7 | 4 of 4 | $5.00 / $30.00 | 1507 | 1499 | 1543 | 1526 |
| 26 | Qwen3.7 Max | Qwen | 64.1 | 3 of 4 | $1.475 / $4.425 | 1517 | 1514 | 1520 | — |
| 27 | Hy3 | Tencent | 63.4 | 3 of 4 | $0.132 / $0.528 | 1522 | 1534 | 1466 | — |
| 28 | GLM 5.1 | Z.ai | 62.9 | 3 of 4 | $0.966 / $3.036 | 1510 | 1506 | 1519 | — |
| 29 | Seed 2.1 Pro Preview | ByteDance | 61.6 | 4 of 4 | — | 1522 | 1524 | 1493 | 1506 |
| 30 | GPT-5.6 Luna (xhigh) | OpenAI | 61.6 | 4 of 4 | $0.10 / $0.60 | 1517 | 1505 | 1537 | 1493 |
| 31 | Kimi K2.6 | MoonshotAI | 61.2 | 4 of 4 | $0.5415 / $2.28 | 1509 | 1504 | 1494 | 1520 |
| 32 | GPT-5.5 (high) | OpenAI | 61.0 | 4 of 4 | $5.00 / $30.00 | 1487 | 1478 | 1540 | 1511 |
| 33 | Claude Opus 4.7 (thinking) (partial coverage) | Anthropic | 60.8 | 1 of 4 | $5.00 / $25.00 | — | — | — | 1579 |
| 34 | Claude Opus 4.5 (high 32K) | Anthropic | 60.4 | 3 of 4 | $5.00 / $25.00 | 1494 | 1480 | 1512 | — |
| 35 | Gemini 3.5 Flash (high) | Google | 58.9 | 4 of 4 | $1.50 / $9.00 | 1499 | 1492 | 1514 | 1492 |
| 36 | Qwen3.6 Max Preview | Qwen | 58.7 | 3 of 4 | $1.027 / $6.162 | 1479 | 1470 | 1496 | — |
| 37 | Gemini 3.5 Flash (medium) | Google | 58.3 | 4 of 4 | $1.50 / $9.00 | 1489 | 1483 | 1521 | 1490 |
| 38 | Claude Opus 4.6 (thinking) (partial coverage) | Anthropic | 57.6 | 1 of 4 | $5.00 / $25.00 | — | — | — | 1540 |
| 39 | MiMo-V2.5-Pro | Xiaomi | 57.5 | 3 of 4 | $0.435 / $0.87 | 1474 | 1462 | 1482 | — |
| 40 | Claude Opus 4.5 | Anthropic | 56.8 | 3 of 4 | $5.00 / $25.00 | 1468 | 1457 | 1482 | — |
| 41 | Kimi K2.7 Code | MoonshotAI | 56.7 | 4 of 4 | $0.71 / $3.50 | 1473 | 1473 | 1461 | 1512 |
| 42 | GPT-5.4 (high) | OpenAI | 56.0 | 3 of 4 | $2.50 / $15.00 | 1463 | 1453 | 1477 | — |
| 43 | GPT-5.5 | OpenAI | 55.8 | 4 of 4 | $5.00 / $30.00 | 1457 | 1445 | 1504 | 1499 |
| 44 | MiniMax M3 | MiniMax | 55.2 | 4 of 4 | $0.30 / $1.20 | 1490 | 1486 | 1463 | 1475 |
| 45 | Gemini 3.6 Flash (partial coverage) | Google | 54.3 | 1 of 4 | $0.75 / $3.75 | — | — | — | 1529 |
| 46 | GPT-5.4 (medium) | OpenAI | 53.9 | 3 of 4 | $2.50 / $15.00 | 1442 | 1429 | 1474 | — |
| 47 | Gemini 3.1 Pro Preview | Google | 53.2 | 4 of 4 | $2.00 / $12.00 | 1447 | 1437 | 1494 | 1481 |
| 48 | DeepSeek V4 Pro | DeepSeek | 52.7 | 3 of 4 | $1.168 / $2.336 | 1446 | 1440 | 1436 | — |
| 49 | Qwen3.6 Plus | Qwen | 52.0 | 4 of 4 | $0.325 / $1.95 | 1460 | 1455 | 1461 | 1470 |
| 50 | GLM 4.7 | Z.ai | 50.9 | 3 of 4 | $0.40 / $1.75 | 1434 | 1429 | 1446 | — |
| 51 | GLM 5 | Z.ai | 50.4 | 3 of 4 | $0.60 / $1.92 | 1436 | 1425 | 1454 | — |
| 52 | MiMo-V2.5 | Xiaomi | 50.0 | 3 of 4 | $0.14 / $0.28 | 1438 | 1427 | 1427 | — |
| 53 | Mimo v2 Pro | Xiaomi | 50.0 | 3 of 4 | — | 1434 | 1426 | 1439 | — |
| 54 | GPT-5 (medium) | OpenAI | 49.5 | 2 of 4 | $1.25 / $10.00 | 1419 | — | 1429 | — |
| 55 | Gemini 3 Flash Preview | Google | 49.1 | 4 of 4 | $0.50 / $3.00 | 1438 | 1429 | 1444 | 1455 |
| 56 | GPT-5.2 | OpenAI | 49.0 | 2 of 4 | $1.75 / $14.00 | 1418 | — | 1428 | — |
| 57 | Gemini 3 Pro | Google | 47.4 | 4 of 4 | — | 1438 | 1396 | 1469 | 1448 |
| 58 | Kimi K2.5 (thinking) | MoonshotAI | 46.2 | 4 of 4 | $0.57 / $2.85 | 1436 | 1429 | 1429 | 1441 |
| 59 | Qwen3.5 397B A17B | Qwen | 45.4 | 3 of 4 | $0.39 / $2.34 | 1400 | 1389 | 1413 | — |
| 60 | GPT-5.1 (medium) | OpenAI | 45.4 | 2 of 4 | $1.25 / $10.00 | 1391 | — | 1401 | — |
| 61 | GPT-5.4 Mini (high) | OpenAI | 45.0 | 3 of 4 | $0.75 / $4.50 | 1397 | 1387 | 1423 | — |
| 62 | GPT-5.3-Codex | OpenAI | 44.6 | 4 of 4 | $1.75 / $14.00 | 1409 | 1398 | 1429 | 1446 |
| 63 | Inkling | Thinking Machines | 44.4 | 3 of 4 | $0.95 / $4.05 | 1405 | 1398 | 1383 | — |
| 64 | Gemini 3.5 Flash Lite | Google | 44.4 | 3 of 4 | $0.30 / $2.50 | 1449 | 1425 | — | 1426 |
| 65 | MiniMax M2.7 | MiniMax | 44.3 | 3 of 4 | $0.30 / $1.20 | 1397 | 1387 | 1401 | — |
| 66 | Claude Opus 4.1 | Anthropic | 44.0 | 2 of 4 | $15.00 / $75.00 | 1389 | — | 1398 | — |
| 67 | Claude Sonnet 4.5 (high 32K) | Anthropic | 43.1 | 3 of 4 | $3.00 / $15.00 | 1392 | 1386 | 1399 | — |
| 68 | MiniMax M2.5 | MiniMax | 42.4 | 3 of 4 | $0.22 / $0.90 | 1384 | 1373 | 1414 | — |
| 69 | GPT-5.4 | OpenAI | 42.2 | 4 of 4 | $2.50 / $15.00 | 1390 | 1389 | 1413 | 1449 |
| 70 | MiniMax M2.1 | MiniMax | 41.5 | 3 of 4 | $0.30 / $1.20 | 1387 | 1369 | 1401 | — |
| 71 | Claude Sonnet 4.5 | Anthropic | 41.1 | 3 of 4 | $3.00 / $15.00 | 1386 | 1384 | 1390 | — |
| 72 | Kimi K2.5 Instant | Moonshot AI | 40.6 | 4 of 4 | — | 1405 | 1391 | 1414 | 1418 |
| 73 | Solar Pro 4 | Upstage | 39.3 | 3 of 4 | $0.03 / $0.12 | 1371 | 1357 | 1393 | — |
| 74 | Grok 4.20 Beta 0309 (reasoning) | xAI | 38.4 | 3 of 4 | — | 1374 | 1367 | 1367 | — |
| 75 | Gemma 4 31B | Google | 38.2 | 3 of 4 | $0.10 / $0.34 | 1364 | 1357 | 1377 | — |
| 76 | Gemini 3 Flash Preview (thinking minimal) | Google | 37.7 | 4 of 4 | $0.50 / $3.00 | 1383 | 1374 | 1398 | 1432 |
| 77 | Gemma 4 26B A4B  | Google | 37.6 | 3 of 4 | $0.12 / $0.40 | 1362 | 1354 | 1372 | — |
| 78 | Muse Glimmer 30B | Meta | 37.6 | 3 of 4 | $0.35 / $1.50 | 1359 | 1352 | 1384 | — |
| 79 | GLM 5V Turbo | Z.ai | 36.9 | 4 of 4 | $1.20 / $4.00 | 1400 | 1384 | 1364 | 1423 |
| 80 | Qwen3.5-27B | Qwen | 36.7 | 3 of 4 | $0.195 / $1.56 | 1358 | 1345 | 1393 | — |
| 81 | DeepSeek V3.2 (thinking) | DeepSeek | 36.6 | 3 of 4 | $0.269 / $0.40 | 1360 | 1350 | 1369 | — |
| 82 | GPT-5.1 (high) (partial coverage) | OpenAI | 36.4 | 1 of 4 | $1.25 / $10.00 | — | — | — | 1420 |
| 83 | GLM 4.6 | Z.ai | 36.3 | 2 of 4 | $0.55 / $2.20 | 1340 | — | 1350 | — |
| 84 | GPT-5.1-Codex | OpenAI | 35.4 | 2 of 4 | $1.25 / $10.00 | 1336 | — | 1346 | — |
| 85 | Qwen3.5-122B-A10B | Qwen | 34.9 | 3 of 4 | $0.29 / $2.40 | 1358 | 1350 | 1357 | — |
| 86 | Hunyuan Hy3 Preview | Tencent | 34.5 | 3 of 4 | — | 1356 | 1348 | 1360 | — |
| 87 | MiniMax M2 | MiniMax | 32.7 | 2 of 4 | $0.255 / $1.02 | 1297 | — | 1307 | — |
| 88 | Laguna M.1 | Poolside | 32.3 | 3 of 4 | — | 1347 | 1338 | 1323 | — |
| 89 | GPT-5.2-Codex | OpenAI | 32.1 | 3 of 4 | $1.75 / $14.00 | 1338 | 1330 | 1348 | — |
| 90 | Mimo v2 Flash | Xiaomi | 31.4 | 3 of 4 | — | 1330 | 1309 | 1353 | — |
| 91 | Grok 4.3 | SpaceXAI | 30.9 | 4 of 4 | $1.25 / $2.50 | 1354 | 1351 | 1362 | 1375 |
| 92 | DeepSeek V3.2 Exp | DeepSeek | 30.9 | 2 of 4 | $0.27 / $0.41 | 1272 | — | 1282 | — |
| 93 | Claude Haiku 4.5 | Anthropic | 30.4 | 3 of 4 | $1.00 / $5.00 | 1326 | 1323 | 1316 | — |
| 94 | Kimi K2 Thinking Turbo | Moonshot AI | 30.2 | 3 of 4 | — | 1323 | 1317 | 1327 | — |
| 95 | DeepSeek V3.2 | DeepSeek | 30.1 | 3 of 4 | $0.269 / $0.40 | 1325 | 1336 | 1304 | — |
| 96 | KAT-Coder-Pro V1 | Kwaipilot | 29.8 | 2 of 4 | — | 1255 | — | 1265 | — |
| 97 | GPT-5.1-Codex-Mini | OpenAI | 28.9 | 2 of 4 | $0.25 / $2.00 | 1244 | — | 1254 | — |
| 98 | GPT-5.1 | OpenAI | 28.2 | 4 of 4 | $1.25 / $10.00 | 1341 | 1305 | 1364 | 1345 |
| 99 | Mimo v2 Flash (thinking) | Xiaomi | 28.0 | 3 of 4 | — | 1292 | 1231 | 1346 | — |
| 100 | Laguna XS.2 | Poolside | 27.6 | 3 of 4 | — | 1303 | 1295 | 1281 | — |
| 101 | Mistral Medium 3.5 | Mistral | 27.5 | 3 of 4 | $1.50 / $7.50 | 1265 | 1255 | 1314 | — |
| 102 | Qwen3 Coder 480B A35b Instruct | Qwen | 27.2 | 3 of 4 | — | 1273 | 1257 | 1286 | — |
| 103 | Grok 4.1 (thinking) | xAI | 26.2 | 2 of 4 | — | 1210 | — | 1220 | — |
| 104 | Qwen3.5-35B-A3B | Qwen | 26.0 | 3 of 4 | $0.225 / $1.80 | 1250 | 1234 | 1298 | — |
| 105 | Grok Code Fast 1 | xAI | 24.6 | 2 of 4 | — | 1164 | — | 1172 | — |
| 106 | Qwen3.5-Flash | Qwen | 24.5 | 3 of 4 | $0.065 / $0.26 | 1238 | 1224 | 1293 | — |
| 107 | Grok 4 Fast (reasoning) | xAI | 24.1 | 2 of 4 | — | 1161 | — | 1170 | — |
| 108 | Devstral Medium 2507 | Mistral AI | 23.7 | 2 of 4 | — | 1080 | — | 1087 | — |
| 109 | Trinity Large Thinking | Arcee AI | 23.6 | 3 of 4 | $0.22 / $0.85 | 1239 | 1233 | 1241 | — |
| 110 | Grok 4.1 Fast (reasoning) | xAI | 23.4 | 3 of 4 | — | 1240 | 1222 | 1252 | — |
| 111 | Mistral Large 3 | Mistral AI | 23.3 | 3 of 4 | — | 1230 | — | 1239 | 1361 |
| 112 | Gemini 3.1 Flash Lite Preview | Google | 21.7 | 4 of 4 | $0.25 / $1.50 | 1254 | 1242 | 1280 | 1336 |
| 113 | Gemini 2.5 Pro | Google | 21.5 | 3 of 4 | $1.25 / $10.00 | 1226 | — | 1235 | 1283 |
| 114 | Granite 4.1 8B | IBM | 21.2 | 3 of 4 | $0.05 / $0.10 | 1192 | 1179 | 1221 | — |
| 115 | Devstral 2 | Mistral AI | 20.6 | 3 of 4 | — | 1194 | 1139 | 1216 | — |
| 116 | Mercury 2 | Inception | 20.2 | 3 of 4 | $0.25 / $0.75 | 1166 | 1156 | 1199 | — |

## How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 2 of the 4 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

## Data sources

- [LMArena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset): CC BY 4.0. Arena ratings by LMArena, from the public leaderboard dataset.