Benchmark Almanac

DesignArena Android (Kotlin)

Claude Opus 5.5 1,397 leads since September 2026
Not in model cards

Models build native Android apps in Kotlin from real user prompts. The apps run in an Android emulator, and users compare them and vote.

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

Kind
Building whole projects
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
46
Models scored on its own leaderboard

Claude Opus 5.5 leads with 1,397

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.

46 models. Scroll sideways for the rest. Bars start at 900.

  1. 1,397 Claude Opus 5.5
  2. 1,292 Muse Spark 1.3
  3. 1,280 Grok 4.6
  4. 1,276 Claude Fable 5
  5. 1,260 Claude Sonnet 5.5
  6. 1,255 Kimi K3
  7. 1,244 Claude Opus 5
  8. 1,233 Gemini 3.7 Flash
  9. 1,227 Grok 4.5
  10. 1,221 Claude Opus 4.8
  11. 1,219 GPT-5.6 Sol
  12. 1,213 Claude Sonnet 5
  13. 1,210 Muse Spark 1.2
  14. 1,202 MiniMax-M3.1-Flash
  15. 1,200 GPT-6.1 Sol
  16. 1,197 Qwen 3.8 Max
  17. 1,193 Gemini 3.6 Flash
  18. 1,189 GPT-6 Astra
  19. 1,182 Claude Sonnet 4.6
  20. 1,181 GPT-6 Luna
  21. 1,180 Claude Opus 4.6
  22. 1,174 Gemini 3.5 Flash
  23. 1,174 Muse Spark 1.1
  24. 1,173 GLM-5.1
  25. 1,170 GPT-6 Sol
  26. 1,169 GLM-5.2
  27. 1,162 Claude Opus 4.5
  28. 1,156 MiniMax-M3
  29. 1,153 GPT-5.5
  30. 1,151 Kimi K2.6
  31. 1,149 Kimi K3.1 Pro
  32. 1,148 Qwen3.7-Max
  33. 1,135 GLM-5
  34. 1,129 Kimi K2.7 Code
  35. 1,105 Kimi K2.5
  36. 1,097 GLM-4.7
  37. 1,086 Gemini 3.8 Flash
  38. 1,072 GLM-4.6
  39. 1,056 GPT-5.6 Terra
  40. 1,045 GPT-5.2
  41. 1,036 Grok 4.20
  42. 1,035 Gemini 3.1 Pro
  43. 1,034 Gemini 3 Flash
  44. 1,021 GPT-5.4
  45. 1,020 Gemini 3 Pro
  46. 955 Grok 4.3

Bars start at 900.

  1. 1 Claude Opus 5.5 1,397
  2. 2 Muse Spark 1.3 xhigh, max 1,292
  3. 3 Grok 4.6 1,280
  4. 4 Claude Fable 5 1,276
  5. 5 Claude Sonnet 5.5 1,260
  6. 6 Kimi K3 1,255
  7. 7 Claude Opus 5 1,244
  8. 8 Gemini 3.7 Flash 1,233
  9. 9 Grok 4.5 1,227
  10. 10 Claude Opus 4.8 1,221
  11. 11 GPT-5.6 Sol medium, xhigh 1,219
  12. 12 Claude Sonnet 5 1,213
  13. 13 Muse Spark 1.2 1,210
  14. 14 MiniMax-M3.1-Flash 1,202
  15. 15 GPT-6.1 Sol 1,200
  16. 16 Qwen 3.8 Max 1,197
  17. 17 Gemini 3.6 Flash 1,193
  18. 18 GPT-6 Astra xhigh 1,189
  19. 19 Claude Sonnet 4.6 1,182
  20. 20 GPT-6 Luna 1,181
  21. 21 Claude Opus 4.6 1,180
  22. 22 Gemini 3.5 Flash 1,174
  23. 23 Muse Spark 1.1 1,174
  24. 24 GLM-5.1 1,173
  25. 25 GPT-6 Sol 1,170
  26. 26 GLM-5.2 1,169
  27. 27 Claude Opus 4.5 1,162
  28. 28 MiniMax-M3 1,156
  29. 29 GPT-5.5 1,153
  30. 30 Kimi K2.6 1,151
  31. 31 Kimi K3.1 Pro 1,149
  32. 32 Qwen3.7-Max 1,148
  33. 33 GLM-5 1,135
  34. 34 Kimi K2.7 Code 1,129
  35. 35 Kimi K2.5 thinking 1,105
  36. 36 GLM-4.7 1,097
  37. 37 Gemini 3.8 Flash 1,086
  38. 38 GLM-4.6 1,072
  39. 39 GPT-5.6 Terra 1,056
  40. 40 GPT-5.2 1,045
  41. 41 Grok 4.20 reasoning 1,036
  42. 42 Gemini 3.1 Pro 1,035
  43. 43 Gemini 3 Flash 1,034
  44. 44 GPT-5.4 1,021
  45. 45 Gemini 3 Pro 1,020
  46. 46 Grok 4.3 955

Source: Official DesignArena Android (Kotlin) leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 5.5 1,397 Sep 2026 Official leaderboard
2 Muse Spark 1.3 1,292 xhigh, max Sep 2026 Official leaderboard
3 Grok 4.6 1,280 Aug 2026 Official leaderboard
4 Claude Fable 5 1,276 Jun 2026 Official leaderboard
5 Claude Sonnet 5.5 1,260 Sep 2026 Official leaderboard
6 Kimi K3 1,255 Jul 2026 Official leaderboard
7 Claude Opus 5 1,244 Jul 2026 Official leaderboard
8 Gemini 3.7 Flash 1,233 Aug 2026 Official leaderboard
9 Grok 4.5 1,227 Jul 2026 Official leaderboard
10 Claude Opus 4.8 1,221 May 2026 Official leaderboard
11 GPT-5.6 Sol 1,219 medium, xhigh Jul 2026 Official leaderboard
12 Claude Sonnet 5 1,213 Jun 2026 Official leaderboard
13 Muse Spark 1.2 1,210 Aug 2026 Official leaderboard
14 MiniMax-M3.1-Flash 1,202 Oct 2026 Official leaderboard
15 GPT-6.1 Sol 1,200 Sep 2026 Official leaderboard
16 Qwen 3.8 Max 1,197 Aug 2026 Official leaderboard
17 Gemini 3.6 Flash 1,193 Jul 2026 Official leaderboard
18 GPT-6 Astra 1,189 xhigh Sep 2026 Official leaderboard
19 Claude Sonnet 4.6 1,182 Feb 2026 Official leaderboard
20 GPT-6 Luna 1,181 Sep 2026 Official leaderboard
21 Claude Opus 4.6 1,180 Feb 2026 Official leaderboard
22 Gemini 3.5 Flash 1,174 May 2026 Official leaderboard
23 Muse Spark 1.1 1,174 Jul 2026 Official leaderboard
24 GLM-5.1 1,173 Apr 2026 Official leaderboard
25 GPT-6 Sol 1,170 Sep 2026 Official leaderboard
26 GLM-5.2 1,169 Jun 2026 Official leaderboard
27 Claude Opus 4.5 1,162 Nov 2025 Official leaderboard
28 MiniMax-M3 1,156 Jun 2026 Official leaderboard
29 GPT-5.5 1,153 Apr 2026 Official leaderboard
30 Kimi K2.6 1,151 Apr 2026 Official leaderboard
31 Kimi K3.1 Pro 1,149 Oct 2026 Official leaderboard
32 Qwen3.7-Max 1,148 May 2026 Official leaderboard
33 GLM-5 1,135 Feb 2026 Official leaderboard
34 Kimi K2.7 Code 1,129 Jun 2026 Official leaderboard
35 Kimi K2.5 1,105 thinking Jan 2026 Official leaderboard
36 GLM-4.7 1,097 Dec 2025 Official leaderboard
37 Gemini 3.8 Flash 1,086 Sep 2026 Official leaderboard
38 GLM-4.6 1,072 Sep 2025 Official leaderboard
39 GPT-5.6 Terra 1,056 Jul 2026 Official leaderboard
40 GPT-5.2 1,045 Dec 2025 Official leaderboard
41 Grok 4.20 1,036 reasoning Feb 2026 Official leaderboard
42 Gemini 3.1 Pro 1,035 Feb 2026 Official leaderboard
43 Gemini 3 Flash 1,034 Dec 2025 Official leaderboard
44 GPT-5.4 1,021 Mar 2026 Official leaderboard
45 Gemini 3 Pro 1,020 Nov 2025 Official leaderboard
46 Grok 4.3 955 Apr 2026 Official leaderboard

9 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.

Z.ai Anthropic xAI Meta
Own leaderboard 9 leaders
Claude Opus 5.5 1,397

Source: Official DesignArena Android (Kotlin) 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
900 1,000 1,100 1,200 1,300 1,400 2026 2027 900 1,000 1,100 1,200 1,300 1,400 2026 2027

Source: Official DesignArena Android (Kotlin) 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