evals.report
BenchmarksLabsCompareRun guidesIn the wild
BenchmarksChat preference

WebDev Arena

A live, community-driven leaderboard where two LLMs compete head-to-head to build interactive web applications from user-submitted prompts, with human votes ranking models by a Bradley-Terry (Elo-like) score.

Chat preferenceEloHigher is better
ScoresAboutRun this benchmark

What is WebDev Arena?

A live, community-driven leaderboard where two LLMs compete head-to-head to build interactive web applications from user-submitted prompts, with human votes ranking models by a Bradley-Terry (Elo-like) score. evals.report tracks reported WebDev Arena scores with the model, source, status, date, and run caveats attached — official leaderboard scores, vendor-reported launches, and clearly labeled community runs.

Top reported WebDev Arena score: Claude Opus 4.7 — 1562 (Elo).

ModelLabScore↓Source modelStatusDate
Claude Opus 4.7Anthropic1562—VerifiedApr 16, 2026Details
Qwen3.7 Max PreviewAlibaba / Qwen1541—VerifiedMay 14, 2026Details
Claude Opus 4.6Anthropic1538—VerifiedFeb 5, 2026Details
GLM-5.1OpenZ.ai1533—VerifiedApr 7, 2026Details
Claude Sonnet 4.6Anthropic1523—VerifiedFeb 17, 2026Details
Kimi K2.6OpenMoonshot AI1518—VerifiedApr 20, 2026Details
Muse SparkMeta1508—VerifiedApr 8, 2026Details
Gemini 3.5 FlashGoogle DeepMind1506—VerifiedMay 19, 2026Details
GPT-5.5OpenAI1505—VerifiedApr 23, 2026Details
Qwen 3.6 Max PreviewAlibaba / Qwen1486—VerifiedApr 20, 2026Details
MiMo-V2.5-ProOpenXiaomi1471—VerifiedApr 22, 2026Details
Claude Opus 4.5Anthropic1467—VerifiedNov 24, 2025Details
DeepSeek V4 ProOpenDeepSeek1464—VerifiedApr 24, 2026Details
Qwen 3.6 PlusAlibaba / Qwen1460—VerifiedApr 2, 2026Details
Gemini 3.1 Pro PreviewGoogle DeepMind1448—VerifiedFeb 19, 2026Details
GLM-4.7OpenZ.ai1440—VerifiedDec 22, 2025Details
MiMo-V2.5OpenXiaomi1440—VerifiedApr 22, 2026Details
Gemini 3 ProGoogle DeepMind1438—VerifiedNov 18, 2025Details
Gemini 3 FlashGoogle DeepMind1437—VerifiedDec 17, 2025Details
GLM-5OpenZ.ai1436—VerifiedFeb 11, 2026Details
Kimi K2.5OpenMoonshot AI1431—VerifiedJan 27, 2026Details
GPT-5.3-CodexOpenAI1407—VerifiedFeb 5, 2026Details
GPT-5.2OpenAI1404—VerifiedDec 11, 2025Details
MiniMax M2.7OpenMiniMax1401—VerifiedMar 18, 2026Details
Grok 4.20 beta reasoningxAI1395—VerifiedMar 9, 2026Details
GPT-5OpenAI1394—VerifiedAug 7, 2025Details
Qwen3.5-397B-A17BOpenAlibaba / Qwen1393—VerifiedFeb 16, 2026Details
MiniMax M2.1OpenMiniMax1392—VerifiedDec 23, 2025Details
GPT-5.1OpenAI1391—VerifiedNov 12, 2025Details
GPT-5.4OpenAI1388—VerifiedMar 5, 2026Details
Claude Sonnet 4.5Anthropic1386—VerifiedSep 29, 2025Details
Claude Opus 4.1Anthropic1386—VerifiedAug 5, 2025Details
MiniMax M2.5OpenMiniMax1382—VerifiedFeb 12, 2026Details
Grok 4.3xAI1377—VerifiedApr 17, 2026Details
GLM-4.6OpenZ.ai1355—VerifiedSep 30, 2025Details
GPT-5.2-CodexOpenAI1335—VerifiedDec 18, 2025Details
DeepSeek V3.2OpenDeepSeek1332—VerifiedDec 1, 2025Details
Kimi K2 ThinkingOpenMoonshot AI1329—VerifiedNov 6, 2025Details
Claude Haiku 4.5Anthropic1322—VerifiedOct 15, 2025Details
Qwen 3 Coder 480BOpenAlibaba / Qwen1282—VerifiedJul 22, 2025Details
Grok 4.1 fast reasoningxAI1234—VerifiedNov 19, 2025Details
Mistral LargeMistral AI1223—VerifiedFeb 26, 2024Details
Gemini 2.5 ProGoogle DeepMind1204—VerifiedMar 25, 2025Details

Each row reports the model’s Elo on WebDev Arena. Click a row for the full run context.