Valumigo

LMArena · Web development

We regularly fetch and display public benchmark data: LMArena user voting rankings, Epoch AI test scores, and OpenRouter usage rankings. Each leaderboard shows both its publication date and the date we retrieved it. These scores are not assigned by this site.

LMArena scores come from people comparing two models' answers side by side and voting for the better one, using an Elo-based system. If score differences fall within the confidence intervals, treat the models as roughly comparable.

How to read this leaderboard

What it measures: A WebDev Arena ranking based on votes anonymously comparing running webpages or web apps generated from the same request.

How to read the score: A relative score reflecting which running results users preferred. It is not an objective functional test pass rate or maintenance score.

Caveats: The evaluation is limited to supported frameworks and generation and execution conditions, so it does not guarantee long-term maintenance capabilities for large projects.

Insights from this ranking

  • The 95% confidence intervals for 1st-place claude-opus-5.5-max and 2nd-place gpt-6-astra-max do not overlap. This aggregation provides relatively clear evidence that the 1st-place model leads.
  • Anthropic has the most models among the top 10, with 5.
  • The 1st-place model received 2,062 votes, and this leaderboard includes 50 models.

Web development LMArena

Published 2026-10-01 · Retrieved 2026-10-04

156016201680174018001860
Anthropicclaude-opus-5.5-max
1815
Ogpt-6-astra-max
1788
Anthropicclaude-sonnet-5.5-xhigh
1786
Ogpt-6.1-sol-max
1758
Anthropicclaude-fable-5.1-max
1749
Anthropicclaude-opus-5-max
1695
Ogpt-6-sol-max
1689
Googlegemini-4-argon-high
1680
Alibabaqwen3.8-max
1671
Alibabaqwen3.8-max-0902
1670
Moonshot AIkimi-k3-max
1658
Metamuse-spark-1.3-max
1657
xgrok-4.7-xhigh
1638
Alibabaqwen3.8-flash-next
1637
Thy4-preview
1633
Anthropicclaude-fable-5-high
1625
Zglm-5.3-max
1623
DeepSeekdeepseek-v4.1-flash-max
1620
Ogpt-5.6-sol-xhigh (codex-harness)
1620
xgrok-4.6-high
1620

Dots show scores; horizontal lines show 95% confidence intervals. Overlapping intervals suggest similar performance. The chart shows only the highest-scoring reasoning setting, such as high or max, for each model. The table shows rankings for all settings.

Top ranking by company

  • AnthropicAnthropicclaude-opus-5.5-max#1
  • OOpenAIgpt-6-astra-max#2
  • GoogleGooglegemini-4-argon-high#9
  • AlibabaAlibabaqwen3.8-max#10
  • Moonshot AIMoonshot AIkimi-k3-max#13
  • MetaMetamuse-spark-1.3-max#14
  • xxAIgrok-4.7-xhigh#15
  • TTencenthy4-preview#17
  • ZZ.aiglm-5.3-max#20
View table (top 50)
RankModelDeveloperScore95% confidence intervalVotes
1#1Anthropicclaude-opus-5.5-maxAnthropic18151799–18312,062
2#2Ogpt-6-astra-maxOpenAI17881777–17986,123
3#3Anthropicclaude-sonnet-5.5-xhighAnthropic17861768–18041,531
4#4Ogpt-6.1-sol-maxOpenAI17581741–17751,620
5#5Anthropicclaude-fable-5.1-maxAnthropic17491740–17596,318
6#6Anthropicclaude-sonnet-5.5-highAnthropic17151702–17292,527
7#7Anthropicclaude-opus-5-maxAnthropic16951689–170116,954
8#8Ogpt-6-sol-maxOpenAI16891678–17003,411
9#9Googlegemini-4-argon-highGoogle16801666–16932,422
10#10Alibabaqwen3.8-maxAlibaba16711659–16833,454
11#11Alibabaqwen3.8-max-0902Alibaba16701662–16789,485
12#12Anthropicclaude-opus-5-highAnthropic16601654–166621,506
13#13Moonshot AIkimi-k3-maxMoonshot AI16581651–166416,513
14#14Metamuse-spark-1.3-maxMeta16571648–16657,249
15#15xgrok-4.7-xhighxAI16381626–16493,216
16#16Alibabaqwen3.8-flash-nextAlibaba16371629–16466,193
17#17Thy4-previewTencent16331624–16435,088
18#18Anthropicclaude-fable-5-highAnthropic16251619–163214,126
19#19Metamuse-spark-1.3 (xHigh)Meta16241616–16336,326
20#20Zglm-5.3-maxZ.ai16231615–16317,658
21#21DeepSeekdeepseek-v4.1-flash-maxDeepSeek16201609–16303,960
22#22Ogpt-5.6-sol-xhigh (codex-harness)OpenAI16201614–162617,128
23#23xgrok-4.6-highxAI16201612–16278,315
24#24Xiaomimimo-v2.6-proXiaomi16181606–16312,734
25#25Zglm-5.3-flashZ.ai16161608–162310,355
26#26Zglm-5.2-maxZ.ai16051599–161114,729
27#27Googlegemini-3.7-flash-highGoogle15921584–15999,624
28#28Alibabaqwen3.8-27bAlibaba15911584–159812,388
29#29Googlegemini-3.8-flash-highGoogle15831575–15918,524
30#30DeepSeekdeepseek-v4-pro-high-20260813DeepSeek15831573–15924,882
31#31DeepSeekdeepseek-v4-flash-highDeepSeek15811572–15905,040
32#32Ogpt-6-luna-maxOpenAI15791570–15876,429
33#33Sstep-5-preview-highStepFun15701557–15832,506
34#34Anthropicclaude-opus-4-7-highAnthropic15571552–156317,982
35#35Anthropicclaude-opus-4-8-highAnthropic15561550–156118,908
36#36Anthropicclaude-opus-4-7Anthropic15551550–156118,174
37#37xgrok-4.5xAI15521545–155811,393
38#38Anthropicclaude-opus-4-6-highAnthropic15461540–155119,519
39#39Metamuse-spark-1.1Meta15421534–15498,070
40#40Anthropicclaude-sonnet-5-highAnthropic15391533–154614,130
41#41Anthropicclaude-opus-4-6Anthropic15371532–154220,762
42#42Googlegemini-3.6-flash-highGoogle15361529–15448,823
43#43Anthropicclaude-opus-4-8Anthropic15341528–154017,880
44#44Metamuse-spark-1.2 (xHigh)Meta15321518–15452,183
45#45Anthropicclaude-sonnet-4-6Anthropic15211516–152722,960
46#46Ogpt-5.6-terra-xhigh (codex-harness)OpenAI15191513–152612,526
47#47Ogpt-5.6-luna-xhigh (codex-harness)OpenAI15181512–152412,776
48#48ByteDanceseed-2.1-pro-previewByteDance15171511–152313,312
49#49Alibabaqwen3.7-max-20260517Alibaba15151508–15228,922
50#50Ogpt-5.5-xhigh (codex-harness)OpenAI15121506–151815,873

Data: LMArena leaderboard dataset (CC BY 4.0). Changes: Top 50 entries per leaderboard; scores are rounded. https://huggingface.co/datasets/lmarena-ai/leaderboard-datasetCompany logos are trademarks of their respective owners and are used only for identification (icons: Simple Icons).Data: Epoch AI — Capabilities & benchmarking (CC BY 4.0). https://epoch.ai/benchmarksSource: OpenRouter (openrouter.ai/rankings), as of 2026-10-04. Licensed under CC BY 4.0. https://openrouter.ai/rankings