Benchmark Almanac

DesignArena React Native

Claude Fable 5.2 1,321 leads since November 2026
Not in model cards

Models build cross-platform mobile apps in React Native from real user prompts. Users compare the running apps and vote for the better one.

DesignArena described the setup in its post Evaluating Mobile App-Building Agents. 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
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
56
Models scored on its own leaderboard

Claude Fable 5.2 leads with 1,321

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.

56 models. Scroll sideways for the rest. Bars start at 1,000.

  1. 1,321 Claude Fable 5.2
  2. 1,296 Claude Opus 5.5
  3. 1,285 GPT-6 Astra
  4. 1,273 Kimi K3.1 Pro
  5. 1,261 Claude Sonnet 5.5
  6. 1,257 Claude Fable 5
  7. 1,256 Kimi K3
  8. 1,256 Claude Fable 5.1
  9. 1,250 Grok 4.7
  10. 1,230 GPT-6.1 Sol
  11. 1,228 GPT-5.6 Sol
  12. 1,228 Gemini 3.7 Flash
  13. 1,228 Gemini 3.8 Flash
  14. 1,226 Grok 4.6
  15. 1,223 Qwen 3.8 Max
  16. 1,221 MiniMax-M3.1-Flash
  17. 1,215 Muse Spark 1.3
  18. 1,210 Claude Opus 4.8
  19. 1,208 Grok 4.5
  20. 1,206 Claude Sonnet 4.6
  21. 1,204 Claude Sonnet 5
  22. 1,202 Claude Opus 4.6
  23. 1,196 GPT-6 Sol
  24. 1,192 Gemini 3.6 Flash
  25. 1,192 Ling-3.1-flash
  26. 1,186 GLM-5.3
  27. 1,183 Claude Opus 4.5
  28. 1,180 GPT-6.1 Terra
  29. 1,176 Muse Spark 1.2
  30. 1,172 Muse Spark 1.1
  31. 1,171 MiniMax-M3
  32. 1,170 Gemini 3.5 Flash
  33. 1,156 GLM-5.2
  34. 1,154 GLM-5.1
  35. 1,152 Qwen3.7-Max
  36. 1,149 Kimi K2.6
  37. 1,141 Kimi K2.5
  38. 1,136 GPT-5.5
  39. 1,136 Kimi K2.7 Code
  40. 1,132 Claude Sonnet 4.5
  41. 1,130 GLM-5
  42. 1,129 GLM-5V-Turbo
  43. 1,119 GPT-6 Luna
  44. 1,099 GLM-4.7
  45. 1,099 GPT-5.6 Terra
  46. 1,090 GLM-4.6
  47. 1,088 GPT-5.2
  48. 1,088 Gemini 3.1 Pro
  49. 1,084 Gemini 3 Flash
  50. 1,084 GPT-5.2 Codex
  51. 1,084 Grok 4.20
  52. 1,081 GPT-5.4
  53. 1,070 Gemini 3 Pro
  54. 1,052 Grok 4.3
  55. 1,048 GPT-5.3 Codex
  56. 1,028 grok-code-fast-1

Bars start at 1,000.

  1. 1 Claude Fable 5.2 1,321
  2. 2 Claude Opus 5.5 1,296
  3. 3 GPT-6 Astra xhigh 1,285
  4. 4 Kimi K3.1 Pro 1,273
  5. 5 Claude Sonnet 5.5 1,261
  6. 6 Claude Fable 5 1,257
  7. 7 Kimi K3 1,256
  8. 8 Claude Fable 5.1 1,256
  9. 9 Grok 4.7 1,250
  10. 10 GPT-6.1 Sol 1,230
  11. 11 GPT-5.6 Sol medium, xhigh 1,228
  12. 12 Gemini 3.7 Flash 1,228
  13. 13 Gemini 3.8 Flash 1,228
  14. 14 Grok 4.6 1,226
  15. 15 Qwen 3.8 Max 1,223
  16. 16 MiniMax-M3.1-Flash 1,221
  17. 17 Muse Spark 1.3 xhigh, max 1,215
  18. 18 Claude Opus 4.8 1,210
  19. 19 Grok 4.5 1,208
  20. 20 Claude Sonnet 4.6 1,206
  21. 21 Claude Sonnet 5 1,204
  22. 22 Claude Opus 4.6 1,202
  23. 23 GPT-6 Sol 1,196
  24. 24 Gemini 3.6 Flash 1,192
  25. 25 Ling-3.1-flash 1,192
  26. 26 GLM-5.3 1,186
  27. 27 Claude Opus 4.5 1,183
  28. 28 GPT-6.1 Terra 1,180
  29. 29 Muse Spark 1.2 1,176
  30. 30 Muse Spark 1.1 1,172
  31. 31 MiniMax-M3 1,171
  32. 32 Gemini 3.5 Flash 1,170
  33. 33 GLM-5.2 1,156
  34. 34 GLM-5.1 1,154
  35. 35 Qwen3.7-Max 1,152
  36. 36 Kimi K2.6 1,149
  37. 37 Kimi K2.5 thinking 1,141
  38. 38 GPT-5.5 1,136
  39. 39 Kimi K2.7 Code 1,136
  40. 40 Claude Sonnet 4.5 1,132
  41. 41 GLM-5 1,130
  42. 42 GLM-5V-Turbo 1,129
  43. 43 GPT-6 Luna 1,119
  44. 44 GLM-4.7 1,099
  45. 45 GPT-5.6 Terra 1,099
  46. 46 GLM-4.6 1,090
  47. 47 GPT-5.2 1,088
  48. 48 Gemini 3.1 Pro 1,088
  49. 49 Gemini 3 Flash 1,084
  50. 50 GPT-5.2 Codex 1,084
  51. 51 Grok 4.20 reasoning 1,084
  52. 52 GPT-5.4 1,081
  53. 53 Gemini 3 Pro 1,070
  54. 54 Grok 4.3 1,052
  55. 55 GPT-5.3 Codex 1,048
  56. 56 grok-code-fast-1 1,028

Source: Official DesignArena React Native leaderboard

#ModelScoreSettingDateSource
1 Claude Fable 5.2 1,321 Nov 2026 Official leaderboard
2 Claude Opus 5.5 1,296 Sep 2026 Official leaderboard
3 GPT-6 Astra 1,285 xhigh Sep 2026 Official leaderboard
4 Kimi K3.1 Pro 1,273 Oct 2026 Official leaderboard
5 Claude Sonnet 5.5 1,261 Sep 2026 Official leaderboard
6 Claude Fable 5 1,257 Jun 2026 Official leaderboard
7 Kimi K3 1,256 Jul 2026 Official leaderboard
8 Claude Fable 5.1 1,256 Sep 2026 Official leaderboard
9 Grok 4.7 1,250 Sep 2026 Official leaderboard
10 GPT-6.1 Sol 1,230 Sep 2026 Official leaderboard
11 GPT-5.6 Sol 1,228 medium, xhigh Jul 2026 Official leaderboard
12 Gemini 3.7 Flash 1,228 Aug 2026 Official leaderboard
13 Gemini 3.8 Flash 1,228 Sep 2026 Official leaderboard
14 Grok 4.6 1,226 Aug 2026 Official leaderboard
15 Qwen 3.8 Max 1,223 Aug 2026 Official leaderboard
16 MiniMax-M3.1-Flash 1,221 Oct 2026 Official leaderboard
17 Muse Spark 1.3 1,215 xhigh, max Sep 2026 Official leaderboard
18 Claude Opus 4.8 1,210 May 2026 Official leaderboard
19 Grok 4.5 1,208 Jul 2026 Official leaderboard
20 Claude Sonnet 4.6 1,206 Feb 2026 Official leaderboard
21 Claude Sonnet 5 1,204 Jun 2026 Official leaderboard
22 Claude Opus 4.6 1,202 Feb 2026 Official leaderboard
23 GPT-6 Sol 1,196 Sep 2026 Official leaderboard
24 Gemini 3.6 Flash 1,192 Jul 2026 Official leaderboard
25 Ling-3.1-flash 1,192 Oct 2026 Official leaderboard
26 GLM-5.3 1,186 Aug 2026 Official leaderboard
27 Claude Opus 4.5 1,183 Nov 2025 Official leaderboard
28 GPT-6.1 Terra 1,180 Oct 2026 Official leaderboard
29 Muse Spark 1.2 1,176 Aug 2026 Official leaderboard
30 Muse Spark 1.1 1,172 Jul 2026 Official leaderboard
31 MiniMax-M3 1,171 Jun 2026 Official leaderboard
32 Gemini 3.5 Flash 1,170 May 2026 Official leaderboard
33 GLM-5.2 1,156 Jun 2026 Official leaderboard
34 GLM-5.1 1,154 Apr 2026 Official leaderboard
35 Qwen3.7-Max 1,152 May 2026 Official leaderboard
36 Kimi K2.6 1,149 Apr 2026 Official leaderboard
37 Kimi K2.5 1,141 thinking Jan 2026 Official leaderboard
38 GPT-5.5 1,136 Apr 2026 Official leaderboard
39 Kimi K2.7 Code 1,136 Jun 2026 Official leaderboard
40 Claude Sonnet 4.5 1,132 Sep 2025 Official leaderboard
41 GLM-5 1,130 Feb 2026 Official leaderboard
42 GLM-5V-Turbo 1,129 Apr 2026 Official leaderboard
43 GPT-6 Luna 1,119 Sep 2026 Official leaderboard
44 GLM-4.7 1,099 Dec 2025 Official leaderboard
45 GPT-5.6 Terra 1,099 Jul 2026 Official leaderboard
46 GLM-4.6 1,090 Sep 2025 Official leaderboard
47 GPT-5.2 1,088 Dec 2025 Official leaderboard
48 Gemini 3.1 Pro 1,088 Feb 2026 Official leaderboard
49 Gemini 3 Flash 1,084 Dec 2025 Official leaderboard
50 GPT-5.2 Codex 1,084 Dec 2025 Official leaderboard
51 Grok 4.20 1,084 reasoning Feb 2026 Official leaderboard
52 GPT-5.4 1,081 Mar 2026 Official leaderboard
53 Gemini 3 Pro 1,070 Nov 2025 Official leaderboard
54 Grok 4.3 1,052 Apr 2026 Official leaderboard
55 GPT-5.3 Codex 1,048 Feb 2026 Official leaderboard
56 grok-code-fast-1 1,028 Aug 2025 Official leaderboard

10 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.

Other labs Anthropic OpenAI
Own leaderboard 10 leaders
Claude Fable 5.2 1,321

Source: Official DesignArena React Native 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 2027 1,000 1,100 1,200 1,300 1,400 2026 2027

Source: Official DesignArena React Native 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