Skip to main content
BenchLM
Data

Design Arena Website Elo (Design Arena Website)

We show this table for reference; we do not rank on it.

Data verified

A display-only Design Arena website-generation Elo score surfaced on OpenRouter model benchmark pages.

Benchmark score on Design Arena Website — November 13, 2026

We compile the Design Arena Website rows from secondary reports. Muse Wonder leads the table at 1367, followed by Muse Spark 1.3 (1351) and Claude Fable 5.2 (1341). We do not use these results to rank models overall.

112 modelsMultimodal & GroundedCurrentDisplay onlyUpdated November 13, 2026

Benchmark score table (112 models)

None of the 112 results on this page has a documented run by the benchmark or an independent evaluator: 1 reported in another publisher’s document and 111 carried from an upstream page, where who ran them is not recorded.

Benchmark score results for Design Arena Website
RankModel / configurationScoreParameters (B)Open / closed
1Muse WonderMetaCarried from an upstream page
1367
Not reportedClosed
2Muse Spark 1.3MetaCarried from an upstream page
1351
Not reportedClosed
3Claude Fable 5.2AnthropicCarried from an upstream page
1341
Not reportedClosed
4Kimi K3Moonshot AICarried from an upstream page
1329
Not reportedOpen
5Claude Opus 5.5AnthropicCarried from an upstream page
1323
Not reportedClosed
6Kimi K3.1 ProMoonshot AICarried from an upstream page
1319
Not reportedOpen
7Claude Sonnet 5.5AnthropicCarried from an upstream page
1313
Not reportedClosed
8Gemini 4 ArgonGoogleCarried from an upstream page
1313
Not reportedClosed
9Muse Spark 1.2MetaCarried from an upstream page
1309
Not reportedClosed
10GPT-6.1 SolOpenAICarried from an upstream page
1308
Not reportedClosed
11Claude Fable 5.1AnthropicCarried from an upstream page
1304
Not reportedClosed
12Claude Opus 5AnthropicCarried from an upstream page
1303
Not reportedClosed
13MiMo-V2.6-ProXiaomiCarried from an upstream page
1303
Not reportedOpen
14Muse Spark OpenMetaCarried from an upstream page
1301
Not reportedOpen
15Gemini 3.7 FlashGoogleCarried from an upstream page
1298
Not reportedClosed
16DeepSeek V4.1 Flash 1017DeepSeekCarried from an upstream page
1296
Not reportedOpen
17Gemini 3.8 FlashGoogleCarried from an upstream page
1295
Not reportedClosed
18Gemini 3.6 FlashGoogleCarried from an upstream page
1293
Not reportedClosed
19Claude Fable 5AnthropicCarried from an upstream page
1290
Not reportedClosed
20GLM-5.1Z.AICarried from an upstream page
1290
Not reportedOpen
21Claude Opus 4.6AnthropicCarried from an upstream page
1286
Not reportedClosed
22Claude Opus 4.6 (Adaptive)AnthropicCarried from an upstream page
1286
Not reportedClosed
23Gemini 4 CarbonGoogleCarried from an upstream page
1285
Not reportedClosed
24GLM-5.3Z.AICarried from an upstream page
1284
Not reportedOpen
25Grok 4.6xAICarried from an upstream page
1282
Not reportedClosed
26Claude Sonnet 4.6AnthropicCarried from an upstream page
1279
Not reportedClosed
27GLM-5.2Z.AICarried from an upstream page
1278
Not reportedOpen
28Grok 4.5xAICarried from an upstream page
1276
Not reportedClosed
29Qwen3.7 MaxAlibabaCarried from an upstream page
1270
Not reportedClosed
30Claude Sonnet 5AnthropicCarried from an upstream page
1268
Not reportedClosed
31Muse Spark 1.1MetaCarried from an upstream page
1265
Not reportedClosed
32GLM-5.3-FlashZ.AICarried from an upstream page
1265
Not reportedOpen
33Gemini 3.5 FlashGoogleCarried from an upstream page
1264
Not reportedClosed
34Kimi K2.6Moonshot AICarried from an upstream page
1264
Not reportedOpen
35Qwen3.7 PlusAlibabaCarried from an upstream page
1263
Not reportedClosed
36MiMo-V2.5-ProXiaomiCarried from an upstream page
1263
Not reportedClosed
37MiMo-V2.5XiaomiCarried from an upstream page
1262
Not reportedClosed
38Kimi K2.7 CodeMoonshot AICarried from an upstream page
1259
Not reportedOpen
39DeepSeek V4 Pro 0813DeepSeekCarried from an upstream page
1258
Not reportedOpen
40DeepSeek V4 Pro (High)DeepSeekCarried from an upstream page
1258
Not reportedOpen
41Claude Haiku 5.5AnthropicCarried from an upstream page
1256
Not reportedClosed
42GLM-5-TurboZ.AICarried from an upstream page
1256
Not reportedClosed
43Claude Opus 4.8AnthropicCarried from an upstream page
1254
Not reportedClosed
44Hy4TencentCarried from an upstream page
1254
Not reportedPending
45Mistral Large 4MistralCarried from an upstream page
1253
Not reportedOpen
46MiniMax M3MiniMaxCarried from an upstream page
1252
Not reportedOpen
47GPT-5.5OpenAICarried from an upstream page
1251
Not reportedClosed
48Gemini 3.1 ProGoogleCarried from an upstream page
1247
Not reportedClosed
49Beam 1.0 501B-A23BReflection AICarried from an upstream page
1247
Not reportedOpen
50Grok 4.7xAICarried from an upstream page
1245
Not reportedClosed
51Kimi K2.5 (Reasoning)Moonshot AICarried from an upstream page
1243
Not reportedOpen
52Claude Opus 4.5AnthropicCarried from an upstream page
1243
Not reportedClosed
53Kimi K2.5Moonshot AICarried from an upstream page
1243
Not reportedOpen
54Claude Opus 4.5 ThinkingAnthropicCarried from an upstream page
1243
Not reportedClosed
55GLM-5Z.AICarried from an upstream page
1242
Not reportedOpen
56GLM-5 (Reasoning)Z.AICarried from an upstream page
1242
Not reportedOpen
57Qwen3.6 PlusAlibabaCarried from an upstream page
1237
Not reportedClosed
58MiniMax M2.7MiniMaxCarried from an upstream page
1237
Not reportedOpen
59DeepSeek V4 FlashDeepSeekCarried from an upstream page
1232
Not reportedOpen
60GLM-5V-TurboZ.AICarried from an upstream page
1224
Not reportedClosed
61Grok 4.20xAICarried from an upstream page
1224
Not reportedClosed
62InklingThinking Machines LabCarried from an upstream page
1220
Not reportedOpen
63GLM-4.7Z.AICarried from an upstream page
1220
Not reportedOpen
64MiniMax M2.5MiniMaxCarried from an upstream page
1218
Not reportedClosed
65GPT-5.4OpenAICarried from an upstream page
1215
Not reportedClosed
66DeepSeek V4 Flash 0731DeepSeekCarried from an upstream page
1202
Not reportedOpen
67DeepSeek V4 Flash (High)DeepSeekCarried from an upstream page
1202
Not reportedOpen
68Gemini 3 FlashGoogleCarried from an upstream page
1192
Not reportedClosed
69Step 3.7 FlashStepFunCarried from an upstream page
1192
Not reportedOpen
70Grok 4.3xAIReported by a third party
1191
Not reportedClosed
71GLM-4.7-FlashZ.AICarried from an upstream page
1190
Not reportedOpen
72GPT-5.2OpenAICarried from an upstream page
1189
Not reportedClosed
73Claude Sonnet 4.5 ThinkingAnthropicCarried from an upstream page
1184
Not reportedClosed
74Claude Sonnet 4.5AnthropicCarried from an upstream page
1184
Not reportedClosed
75GPT-5.1OpenAICarried from an upstream page
1183
Not reportedClosed
76Muse Glimmer 30BMetaCarried from an upstream page
1180
Not reportedOpen
77GPT-5 (high)OpenAICarried from an upstream page
1179
Not reportedClosed
78GPT-5 (medium)OpenAICarried from an upstream page
1179
Not reportedClosed
79Hy3TencentCarried from an upstream page
1176
Not reportedOpen
80Solar Pro 4UpstageCarried from an upstream page
1172
Not reportedClosed
81Claude 4.1 OpusAnthropicCarried from an upstream page
1171
Not reportedClosed
82DeepSeek V3.2DeepSeekCarried from an upstream page
1171
Not reportedOpen
83DeepSeek V3.2 (Thinking)DeepSeekCarried from an upstream page
1171
Not reportedOpen
84GLM-4.5Z.AICarried from an upstream page
1164
Not reportedClosed
85Gemini 2.5 ProGoogleCarried from an upstream page
1163
Not reportedClosed
86GPT-5.1-CodexOpenAICarried from an upstream page
1158
Not reportedClosed
87GPT-5.3 CodexOpenAICarried from an upstream page
1157
Not reportedClosed
88Qwen3-Coder 480B-A35BAlibabaCarried from an upstream page
1153
Not reportedOpen
89DeepSeek-R1-0528DeepSeekCarried from an upstream page
1145
Not reportedOpen
90Nemotron 3 UltraNVIDIACarried from an upstream page
1141
Not reportedOpen
91GLM-4.5-AirZ.AICarried from an upstream page
1141
Not reportedClosed
92GPT-5 miniOpenAICarried from an upstream page
1120
Not reportedClosed
93DeepSeek V3.1 (Reasoning)DeepSeekCarried from an upstream page
1118
Not reportedOpen
94DeepSeek V3.1DeepSeekCarried from an upstream page
1118
Not reportedOpen
95Claude Haiku 4.5AnthropicCarried from an upstream page
1118
Not reportedClosed
96Claude Haiku 4.5 ThinkingAnthropicCarried from an upstream page
1118
Not reportedClosed
97DeepSeek V3DeepSeekCarried from an upstream page
1116
Not reportedOpen
98Gemini 2.5 FlashGoogleCarried from an upstream page
1111
Not reportedClosed
99GPT-5 nanoOpenAICarried from an upstream page
1097
Not reportedClosed
100Mistral Medium 3MistralCarried from an upstream page
1075
Not reportedClosed
101Kimi K2Moonshot AICarried from an upstream page
1046
Not reportedClosed
102GPT-4.1OpenAICarried from an upstream page
1033
Not reportedClosed
103o3OpenAICarried from an upstream page
1032
Not reportedClosed
104GPT-4.1 miniOpenAICarried from an upstream page
993
Not reportedClosed
105o4-mini (high)OpenAICarried from an upstream page
981
Not reportedClosed
106o4-miniOpenAICarried from an upstream page
981
Not reportedClosed
107GPT-4.1 nanoOpenAICarried from an upstream page
969
Not reportedClosed
108GPT-OSS 120BOpenAICarried from an upstream page
964
Not reportedOpen
109Llama 4 MaverickMetaCarried from an upstream page
867
Not reportedOpen
110GPT-OSS 20BOpenAICarried from an upstream page
848
Not reportedOpen
111GPT-4oOpenAICarried from an upstream page
827
Not reportedClosed
112Llama 4 ScoutMetaCarried from an upstream page
746
Not reportedOpen
  • Lab-reported
  • Benchmark-run
  • Independent
What the labels mean

Among the reported Design Arena Website rows, Muse Wonder is first at 1367. The third row is 26 score units behind. The broader top-10 range is 59 score units, so the table still separates the published systems.

112 models have been evaluated on Design Arena Website. The benchmark falls in the Multimodal & Grounded category. Design Arena Website is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Design Arena Website

Year

2026

Tasks

Website generation comparisons

Format

Elo

Difficulty

Design and website generation

OpenRouter's Grok 4.3 benchmark page reports Website at 1294 Elo, 56.5% win rate, 166.3s average generation time, and Top 13%. BenchLM stores the Elo as the benchmark value and keeps the supporting fields in the description.

Freshness and provenance

Version

Design Arena Website 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does Design Arena Website measure?

A display-only Design Arena website-generation Elo score surfaced on OpenRouter model benchmark pages.

Which model scores highest on Design Arena Website?

Muse Wonder by Meta currently leads with a score of 1367 on Design Arena Website.

How many models are evaluated on Design Arena Website?

112 AI models have published results on Design Arena Website in the BenchLM catalog.

Last updated: November 13, 2026 · BenchLM version Design Arena Website 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 7,000+ subscribers.

One email each week. Unsubscribe anytime.