Benchmark Almanac

DesignArena Image to Frontend

Gemini 4 Argon 1,314 leads since September 2026
Not in model cards

Coding agents turn a reference image into a multi-file frontend web app. Users compare the running apps and vote for the better one.

One of 40 DesignArena boards. DesignArena ranks entries with Bradley-Terry ratings from user votes.

Kind
Web and app front ends
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Strongest model tested
GPT-6 Astra #3 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
32
Entries on its own leaderboard, 31 models we track

Gemini 4 Argon leads with 1,314

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

32 entries. Scroll sideways for the rest. Bars start at 1,000.

  1. 1,314 Gemini 4 Argon
  2. 1,292 Claude Opus 5.5
  3. 1,264 Claude Fable 5
  4. 1,260 Claude Fable 5.1
  5. 1,256 Muse Wonder
  6. 1,252 GPT-6 Astra
  7. 1,249 Kimi K3
  8. 1,249 Claude Sonnet 5.5
  9. 1,245 Muse Spark 1.3
  10. 1,238 Ling 3.2 Flash
  11. 1,235 GPT-5.6 Sol
  12. 1,235 DeepSeek V4.1 Flash
  13. 1,233 Muse Spark 1.2
  14. 1,216 GPT-6.1 Sol
  15. 1,205 Grok 4.6
  16. 1,196 Claude Opus 4.8
  17. 1,195 Muse Spark 1.1
  18. 1,191 Grok 4.5
  19. 1,182 Claude Sonnet 5
  20. 1,167 Gemini 3.6 Flash
  21. 1,156 MiniMax-M3
  22. 1,149 Kimi K2.7 Code
  23. 1,147 Claude Sonnet 4.6
  24. 1,139 Gemini 3.5 Flash
  25. 1,128 Grok 4.3
  26. 1,117 Gemini 3.1 Pro
  27. 1,112 Kimi K2.5
  28. 1,109 GLM-5V-Turbo
  29. 1,107 GPT-5.5
  30. 1,104 Gemini 3 Pro
  31. 1,081 GPT-5.6 Terra
  32. 1,059 Gemini 3 Flash

Bars start at 1,000.

  1. 1 Gemini 4 Argon 1,314
  2. 2 Claude Opus 5.5 1,292
  3. 3 Claude Fable 5 1,264
  4. 4 Claude Fable 5.1 1,260
  5. 5 Muse Wonder max 1,256
  6. 6 GPT-6 Astra xhigh 1,252
  7. 7 Kimi K3 1,249
  8. 8 Claude Sonnet 5.5 1,249
  9. 9 Muse Spark 1.3 xhigh 1,245
  10. 10 Ling 3.2 Flash 1,238
  11. 11 GPT-5.6 Sol medium, xhigh 1,235
  12. 12 DeepSeek V4.1 Flash 1017 1,235
  13. 13 Muse Spark 1.2 1,233
  14. 14 GPT-6.1 Sol 1,216
  15. 15 Grok 4.6 1,205
  16. 16 Claude Opus 4.8 1,196
  17. 17 Muse Spark 1.1 1,195
  18. 18 Grok 4.5 1,191
  19. 19 Claude Sonnet 5 1,182
  20. 20 Gemini 3.6 Flash 1,167
  21. 21 MiniMax-M3 1,156
  22. 22 Kimi K2.7 Code 1,149
  23. 23 Claude Sonnet 4.6 1,147
  24. 24 Gemini 3.5 Flash 1,139
  25. 25 Grok 4.3 1,128
  26. 26 Gemini 3.1 Pro 1,117
  27. 27 Kimi K2.5 thinking 1,112
  28. 28 GLM-5V-Turbo 1,109
  29. 29 GPT-5.5 1,107
  30. 30 Gemini 3 Pro 1,104
  31. 31 GPT-5.6 Terra 1,081
  32. 32 Gemini 3 Flash 1,059

Source: Official DesignArena Image to Frontend leaderboard

#EntryScoreSettingDateSource
1 Gemini 4 Argon 1,314 Sep 2026 Official leaderboard
2 Claude Opus 5.5 1,292 Sep 2026 Official leaderboard
3 Claude Fable 5 1,264 Jun 2026 Official leaderboard
4 Claude Fable 5.1 1,260 Sep 2026 Official leaderboard
5 Muse Wonder 1,256 max Oct 2026 Official leaderboard
6 GPT-6 Astra 1,252 xhigh Sep 2026 Official leaderboard
7 Kimi K3 1,249 Jul 2026 Official leaderboard
8 Claude Sonnet 5.5 1,249 Sep 2026 Official leaderboard
9 Muse Spark 1.3 1,245 xhigh Sep 2026 Official leaderboard
10 Ling 3.2 Flash 1,238 Official leaderboard
11 GPT-5.6 Sol 1,235 medium, xhigh Jul 2026 Official leaderboard
12 DeepSeek V4.1 Flash 1,235 1017 Sep 2026 Official leaderboard
13 Muse Spark 1.2 1,233 Aug 2026 Official leaderboard
14 GPT-6.1 Sol 1,216 Sep 2026 Official leaderboard
15 Grok 4.6 1,205 Aug 2026 Official leaderboard
16 Claude Opus 4.8 1,196 May 2026 Official leaderboard
17 Muse Spark 1.1 1,195 Jul 2026 Official leaderboard
18 Grok 4.5 1,191 Jul 2026 Official leaderboard
19 Claude Sonnet 5 1,182 Jun 2026 Official leaderboard
20 Gemini 3.6 Flash 1,167 Jul 2026 Official leaderboard
21 MiniMax-M3 1,156 Jun 2026 Official leaderboard
22 Kimi K2.7 Code 1,149 Jun 2026 Official leaderboard
23 Claude Sonnet 4.6 1,147 Feb 2026 Official leaderboard
24 Gemini 3.5 Flash 1,139 May 2026 Official leaderboard
25 Grok 4.3 1,128 Apr 2026 Official leaderboard
26 Gemini 3.1 Pro 1,117 Feb 2026 Official leaderboard
27 Kimi K2.5 1,112 thinking Jan 2026 Official leaderboard
28 GLM-5V-Turbo 1,109 Apr 2026 Official leaderboard
29 GPT-5.5 1,107 Apr 2026 Official leaderboard
30 Gemini 3 Pro 1,104 Nov 2025 Official leaderboard
31 GPT-5.6 Terra 1,081 Jul 2026 Official leaderboard
32 Gemini 3 Flash 1,059 Dec 2025 Official leaderboard

7 models have held number one on the official leaderboard since 2023

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind Moonshot AI Anthropic
Own leaderboard 7 leaders
Gemini 4 Argon 1,314

Source: Official DesignArena Image to Frontend leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
1,000 1,100 1,200 1,300 1,400 2026 1,000 1,100 1,200 1,300 1,400 2026

Source: Official DesignArena Image to Frontend leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Coding

All 282 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471