Skip to content
Bencher

TestsBenchmarks · Human preference

Design Arena (fullstack)

Crowd-voted Elo for full-stack web apps built by models.

Elo ratingHigher is betterThe test's websiteOfficial page ↗

The best 15 · 19 models tested

Top 15 · best setting per model · 19 models, 21 results (21 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported
Design Arena (fullstack) leaderboard
Thinking levelSetting
1GPT-6 AstraOpenAI1355Defaultgpt-6-astraIndependent testIndependentDesign Arena ↗—1 Oct 2026
2Claude Fable 5.1Anthropic1324Defaultclaude-fable-5-1Independent testIndependentDesign Arena ↗—1 Oct 2026
3Kimi K3Moonshot AI (Kimi)1314Defaultkimi-k3Independent testIndependentDesign Arena ↗—1 Oct 2026
4Muse Spark 1.3Meta (Meta Superintelligence Labs, Muse)1310Maxmuse-spark-1.3-maxIndependent testIndependentDesign Arena ↗—1 Oct 2026
5Claude Opus 5Anthropic1304Defaultclaude-opus-5Independent testIndependentDesign Arena ↗—1 Oct 2026
6Qwen3.8 Max (0902)Qwen (Alibaba)1304Defaultqwen3.8-maxIndependent testIndependentDesign Arena ↗—1 Oct 2026
7Claude Fable 5Anthropic1278Defaultclaude-fable-5Independent testIndependentDesign Arena ↗—1 Oct 2026
8GPT-5.6 SolOpenAI1259Extra highgpt-5.6-sol-xhighIndependent testIndependentDesign Arena ↗—1 Oct 2026
9Claude Opus 4.8Anthropic1247Defaultclaude-opus-4-8Independent testIndependentDesign Arena ↗—1 Oct 2026
10Claude Sonnet 5Anthropic1247Defaultclaude-sonnet-5Independent testIndependentDesign Arena ↗—1 Oct 2026
11Gemini 3.8 FlashGoogle (Gemini / DeepMind)1242Defaultgemini-3.8-flashIndependent testIndependentDesign Arena ↗—1 Oct 2026
12GLM-5.2Z.ai (Zhipu)1239Defaultglm-5.2Independent testIndependentDesign Arena ↗—1 Oct 2026
13GPT-6 SolOpenAI1237Defaultgpt-6-solIndependent testIndependentDesign Arena ↗—1 Oct 2026
14Claude Opus 4.6Anthropic1218Defaultclaude-opus-4-6Independent testIndependentDesign Arena ↗—1 Oct 2026
15Claude Sonnet 4.6Anthropic1211Defaultclaude-sonnet-4-6Independent testIndependentDesign Arena ↗—1 Oct 2026
16Muse Spark 1.2Meta (Meta Superintelligence Labs, Muse)1208Defaultmuse-spark-1.2Independent testIndependentDesign Arena ↗—1 Oct 2026
17Gemini 3.7 FlashGoogle (Gemini / DeepMind)1207Defaultgemini-3.7-flashIndependent testIndependentDesign Arena ↗—1 Oct 2026
18MiniMax M3MiniMax1196Defaultminimax-m3Independent testIndependentDesign Arena ↗—1 Oct 2026
19Gemini 3.6 FlashGoogle (Gemini / DeepMind)1179Defaultgemini-3.6-flashIndependent testIndependentDesign Arena ↗—1 Oct 2026