Benchmark Almanac

DesignArena UI Component

GPT-6 Astra 1,401 leads since September 2026
Not in model cards

Models build one interface component as a single HTML file from a real user prompt. Users compare the rendered components and vote for the better one.

It is one of the five categories behind the Code overall board, with its own prompts on the DesignArena home page. 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
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
164
Entries on its own leaderboard, 139 models we track

GPT-6 Astra leads with 1,401

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. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

164 entries. Scroll sideways for the rest. Bars start at 600.

  1. 1,401 GPT-6 Astra
  2. 1,388 Kimi K3.1 Pro
  3. 1,386 Claude Sonnet 5.5
  4. 1,379 Claude Opus 5.5
  5. 1,368 Muse Wonder
  6. 1,358 GPT-6.1 Sol
  7. 1,357 Claude Fable 5.2
  8. 1,354 Muse Spark 1.3
  9. 1,351 Ling-3.1-flash
  10. 1,349 Gemini 4 Carbon
  11. 1,341 Claude Opus 5
  12. 1,339 Kimi K3
  13. 1,336 Qwen 3.8 Max
  14. 1,336 Gemini 4 Argon
  15. 1,333 Claude Fable 5.1
  16. 1,333 MiMo-V2.6 Pro
  17. 1,327 GPT-5.6 Sol
  18. 1,314 Muse Spark Open
  19. 1,312 Muse Ember
  20. 1,311 Claude Fable 5
  21. 1,306 Gemini 3.8 Flash
  22. 1,304 DeepSeek V4.1 Flash
  23. 1,301 Muse Spark 1.2
  24. 1,298 GLM-5.3-Flash
  25. 1,293 MiniMax-M3.1-Flash
  26. 1,292 GPT-6 Luna
  27. 1,291 Gemini 3.6 Flash
  28. 1,289 GPT-6.1 Terra
  29. 1,288 GLM-5.2
  30. 1,288 Gemini 3.7 Flash
  31. 1,285 Grok 4.6
  32. 1,285 GPT-6 Sol
  33. 1,284 GLM-5.3
  34. 1,281 Muse Spark 1.1
  35. 1,279 Mistral Large 4
  36. 1,275 Claude Opus 4.6
  37. 1,274 Grok 4.7
  38. 1,274 Claude Haiku 5.5
  39. 1,271 Gemini 3 Pro
  40. 1,270 Grok 4.5
  41. 1,269 Claude Sonnet 5
  42. 1,266 Gemini 3.5 Flash
  43. 1,264 Qwen3.7-Max
  44. 1,263 Claude Sonnet 4.6
  45. 1,262 Deca 2.5 Mini
  46. 1,260 GLM-5.1
  47. 1,259 Beam 1.0
  48. 1,258 Kimi K2.6
  49. 1,256 Qwen3.7-Plus
  50. 1,254 GLM-5-Turbo
  51. 1,252 Kimi K2.7 Code
  52. 1,251 Deca 2.6 Ultra
  53. 1,250 GPT-5.5
  54. 1,250 MiMo-V2.5
  55. 1,250 GPT-5.6 Terra
  56. 1,249 Claude Opus 4.8
  57. 1,244 MiMo-V2.5-Pro
  58. 1,241 GPT-5.6 Luna
  59. 1,236 MiniMax-M3
  60. 1,235 Muse Spark
  61. 1,231 GPT-5.4
  62. 1,229 Deca 2.6 Pro
  63. 1,229 Claude Opus 4.5
  64. 1,228 GPT-5.1-Codex
  65. 1,228 Kimi K2.5
  66. 1,228 DeepSeek V4 Flash 0731
  67. 1,226 Mercury 3
  68. 1,223 Qwen 3.6 Plus
  69. 1,222 GLM-5
  70. 1,216 Gemini 3.1 Pro
  71. 1,214 MiniMax-M2.1
  72. 1,212 DeepSeek-V4-Pro
  73. 1,211 Nex N2 Pro
  74. 1,206 Deca 2.5 Ultra
  75. 1,203 MiniMax-M2.7
  76. 1,202 GLM-5V-Turbo
  77. 1,199 glm-4.7-flash
  78. 1,198 Muse Glimmer 30B
  79. 1,195 Tencent Hy3
  80. 1,194 Tencent Hy4
  81. 1,192 Grok 4.20 Beta
  82. 1,190 GLM-4.7
  83. 1,190 Inkling
  84. 1,186 Grok 4.3
  85. 1,181 GPT-5.2
  86. 1,173 DeepSeek-V3.1
  87. 1,171 GPT-5
  88. 1,169 Qwen3.5 Plus 02-15
  89. 1,168 Step 3.7 Flash
  90. 1,164 MiniMax-M2.5
  91. 1,163 Claude Sonnet 4.5
  92. 1,160 Flint
  93. 1,158 DeepSeek-V3.2-Exp
  94. 1,157 GPT-5.1
  95. 1,156 Claude Opus 4.1
  96. 1,155 v0-1.5-lg
  97. 1,155 DeepSeek-V4-Flash
  98. 1,154 Qwen3.5 397B-A17B
  99. 1,154 Nemotron 3.5 Super VL
  100. 1,144 Claude Opus 4
  101. 1,142 Claude 3.7 Sonnet
  102. 1,141 Solar Pro 4
  103. 1,140 AGI-01 Swift
  104. 1,140 DeepSeek-V3.2
  105. 1,139 GLM-4.6
  106. 1,135 GLM-4.5
  107. 1,135 MiMo-V2-Flash
  108. 1,131 Gemini 2.5 Pro
  109. 1,128 MiniMax M2 Stable
  110. 1,128 GPT-5.3 Codex
  111. 1,128 Nemotron 3 Ultra
  112. 1,120 Claude Sonnet 4
  113. 1,114 GLM-4.5-Air
  114. 1,104 Qwen3 Coder 480B A35B Instruct
  115. 1,097 GPT-5 mini (Default)
  116. 1,094 DeepSeek-R1
  117. 1,093 Mistral Large 3
  118. 1,093 Solar Mini 4
  119. 1,091 Mistral Medium 3.1
  120. 1,091 Claude Haiku 4.5
  121. 1,084 DeepSeek-V3
  122. 1,084 Gemini 2.5 Flash
  123. 1,077 AesCoder-4B
  124. 1,072 GPT-5.1-Codex-mini
  125. 1,070 Mistral Medium 3.5
  126. 1,069 Qwen3-Max
  127. 1,068 Prime Intellect: INTELLECT-3
  128. 1,063 Grok 4 Fast
  129. 1,063 Gemini 3.1 Flash-Lite
  130. 1,058 GPT-5 nano (Default)
  131. 1,044 Grok 4.1 Fast
  132. 1,037 Qwen3-Coder-30B-A3B-Instruct
  133. 1,036 Trinity Large Thinking
  134. 1,028 Grok 3
  135. 1,021 Kimi K2
  136. 1,016 Mistral Medium 3 (2505)
  137. 1,016 Gemini 2.5 Flash-Lite
  138. 1,014 Magistral Medium 1.2 (2509)
  139. 1,008 Codestral 2508
  140. 1,008 o3
  141. 1,001 v0-1.5-md
  142. 991 GPT-4.1
  143. 981 grok-code-fast-1
  144. 977 Grok 4
  145. 974 o4-mini
  146. 970 Devstral Medium
  147. 970 Mercury 2
  148. 956 Qwen3-235B-A22B-Instruct-2507
  149. 952 GPT-4.1 mini
  150. 950 Qwen3-235B-A22B
  151. 937 Qwen3-30B-A3B
  152. 936 Qwen3-235B-A22B-Thinking
  153. 921 Magistral Small 1.2 (2509)
  154. 918 gpt-oss-120b
  155. 907 GPT-4.1 nano
  156. 898 Mistral Small 3.2
  157. 890 Llama 4 Maverick
  158. 881 Grok-3 mini
  159. 876 GPT-4o
  160. 850 Olmo 3.1 32B Think
  161. 843 Codestral 2 (2501)
  162. 830 Devstral Small 1.1
  163. 815 Mistral Large 2.1 (2411)
  164. 759 Llama 4 Scout

Bars start at 600.

  1. 1 GPT-6 Astra xhigh 1,401
  2. 2 Kimi K3.1 Pro 1,388
  3. 3 Claude Sonnet 5.5 1,386
  4. 4 Claude Opus 5.5 1,379
  5. 5 Muse Wonder max 1,368
  6. 6 GPT-6.1 Sol 1,358
  7. 7 Claude Fable 5.2 1,357
  8. 8 Muse Spark 1.3 xhigh, max 1,354
  9. 9 Ling-3.1-flash 1,351
  10. 10 Gemini 4 Carbon 1,349
  11. 11 Claude Opus 5 1,341
  12. 12 Kimi K3 1,339
  13. 13 Qwen 3.8 Max 1,336
  14. 14 Gemini 4 Argon 1,336
  15. 15 Claude Fable 5.1 1,333
  16. 16 MiMo-V2.6 Pro 1,333
  17. 17 GPT-5.6 Sol medium, xhigh 1,327
  18. 18 Muse Spark Open xhigh 1,314
  19. 19 Muse Ember 1,312
  20. 20 Claude Fable 5 1,311
  21. 21 Gemini 3.8 Flash 1,306
  22. 22 DeepSeek V4.1 Flash 1,304
  23. 23 Muse Spark 1.2 1,301
  24. 24 GLM-5.3-Flash 1,298
  25. 25 MiniMax-M3.1-Flash 1,293
  26. 26 GPT-6 Luna 1,292
  27. 27 Gemini 3.6 Flash 1,291
  28. 28 GPT-6.1 Terra 1,289
  29. 29 GLM-5.2 1,288
  30. 30 Gemini 3.7 Flash 1,288
  31. 31 Grok 4.6 1,285
  32. 32 GPT-6 Sol 1,285
  33. 33 GLM-5.3 1,284
  34. 34 Muse Spark 1.1 1,281
  35. 35 Mistral Large 4 1,279
  36. 36 Claude Opus 4.6 1,275
  37. 37 Grok 4.7 1,274
  38. 38 Claude Haiku 5.5 1,274
  39. 39 Gemini 3 Pro 1,271
  40. 40 Grok 4.5 1,270
  41. 41 Claude Sonnet 5 1,269
  42. 42 Gemini 3.5 Flash 1,266
  43. 43 Qwen3.7-Max 1,264
  44. 44 Claude Sonnet 4.6 1,263
  45. 45 Deca 2.5 Mini 1,262
  46. 46 GLM-5.1 1,260
  47. 47 Beam 1.0 max 1,259
  48. 48 Kimi K2.6 1,258
  49. 49 Qwen3.7-Plus 1,256
  50. 50 GLM-5-Turbo 1,254
  51. 51 Kimi K2.7 Code 1,252
  52. 52 Deca 2.6 Ultra 1,251
  53. 53 GPT-5.5 1,250
  54. 54 MiMo-V2.5 1,250
  55. 55 GPT-5.6 Terra 1,250
  56. 56 Claude Opus 4.8 1,249
  57. 57 MiMo-V2.5-Pro 1,244
  58. 58 GPT-5.6 Luna 1,241
  59. 59 MiniMax-M3 1,236
  60. 60 Muse Spark 1,235
  61. 61 GPT-5.4 medium 1,231
  62. 62 Deca 2.6 Pro 1,229
  63. 63 Claude Opus 4.5 1,229
  64. 64 GPT-5.1-Codex 1,228
  65. 65 Kimi K2.5 thinking 1,228
  66. 66 DeepSeek V4 Flash 0731 1,228
  67. 67 Mercury 3 1,226
  68. 68 Qwen 3.6 Plus 1,223
  69. 69 GLM-5 1,222
  70. 70 Gemini 3.1 Pro 1,216
  71. 71 MiniMax-M2.1 1,214
  72. 72 DeepSeek-V4-Pro 1,212
  73. 73 Nex N2 Pro 1,211
  74. 74 Deca 2.5 Ultra 2 1,206
  75. 75 MiniMax-M2.7 1,203
  76. 76 GLM-5V-Turbo 1,202
  77. 77 glm-4.7-flash 1,199
  78. 78 Muse Glimmer 30B xhigh 1,198
  79. 79 Tencent Hy3 1,195
  80. 80 Tencent Hy4 1,194
  81. 81 Grok 4.20 Beta reasoning 1,192
  82. 82 GLM-4.7 1,190
  83. 83 Inkling 1,190
  84. 84 Grok 4.3 1,186
  85. 85 GPT-5.2 medium 1,181
  86. 86 DeepSeek-V3.1 thinking 1,173
  87. 87 GPT-5 minimal, high 1,171
  88. 88 Qwen3.5 Plus 02-15 1,169
  89. 89 Step 3.7 Flash 1,168
  90. 90 MiniMax-M2.5 1,164
  91. 91 Claude Sonnet 4.5 thinking 1,163
  92. 92 Flint 1,160
  93. 93 DeepSeek-V3.2-Exp 1,158
  94. 94 GPT-5.1 medium 1,157
  95. 95 Claude Opus 4.1 thinking 1,156
  96. 96 v0-1.5-lg 1,155
  97. 97 DeepSeek-V4-Flash 1,155
  98. 98 Qwen3.5 397B-A17B 1,154
  99. 99 Nemotron 3.5 Super VL 1,154
  100. 100 Claude Opus 4 1,144
  101. 101 Claude 3.7 Sonnet 1,142
  102. 102 Solar Pro 4 1,141
  103. 103 AGI-01 Swift 1,140
  104. 104 DeepSeek-V3.2 1,140
  105. 105 GLM-4.6 1,139
  106. 106 GLM-4.5 1,135
  107. 107 MiMo-V2-Flash 1,135
  108. 108 Gemini 2.5 Pro Jun 2025 1,131
  109. 109 MiniMax M2 Stable 1,128
  110. 110 GPT-5.3 Codex 1,128
  111. 111 Nemotron 3 Ultra 1,128
  112. 112 Claude Sonnet 4 1,120
  113. 113 GLM-4.5-Air 1,114
  114. 114 Qwen3 Coder 480B A35B Instruct 1,104
  115. 115 GPT-5 mini (Default) 1,097
  116. 116 DeepSeek-R1 May 2025 1,094
  117. 117 Mistral Large 3 1,093
  118. 118 Solar Mini 4 1,093
  119. 119 Mistral Medium 3.1 1,091
  120. 120 Claude Haiku 4.5 1,091
  121. 121 DeepSeek-V3 Mar 2025 1,084
  122. 122 Gemini 2.5 Flash Sep 2025 1,084
  123. 123 AesCoder-4B 1,077
  124. 124 GPT-5.1-Codex-mini 1,072
  125. 125 Mistral Medium 3.5 1,070
  126. 126 Qwen3-Max 1,069
  127. 127 Prime Intellect: INTELLECT-3 1,068
  128. 128 Grok 4 Fast 1,063
  129. 129 Gemini 3.1 Flash-Lite 1,063
  130. 130 GPT-5 nano (Default) 1,058
  131. 131 Grok 4.1 Fast reasoning 1,044
  132. 132 Qwen3-Coder-30B-A3B-Instruct 1,037
  133. 133 Trinity Large Thinking 1,036
  134. 134 Grok 3 1,028
  135. 135 Kimi K2 Sep 2025 1,021
  136. 136 Mistral Medium 3 (2505) 1,016
  137. 137 Gemini 2.5 Flash-Lite Sep 2025 1,016
  138. 138 Magistral Medium 1.2 (2509) 1,014
  139. 139 Codestral 2508 1,008
  140. 140 o3 1,008
  141. 141 v0-1.5-md 1,001
  142. 142 GPT-4.1 991
  143. 143 grok-code-fast-1 981
  144. 144 Grok 4 977
  145. 145 o4-mini 974
  146. 146 Devstral Medium 970
  147. 147 Mercury 2 970
  148. 148 Qwen3-235B-A22B-Instruct-2507 956
  149. 149 GPT-4.1 mini 952
  150. 150 Qwen3-235B-A22B 950
  151. 151 Qwen3-30B-A3B 937
  152. 152 Qwen3-235B-A22B-Thinking Jul 2025 936
  153. 153 Magistral Small 1.2 (2509) 921
  154. 154 gpt-oss-120b 918
  155. 155 GPT-4.1 nano 907
  156. 156 Mistral Small 3.2 898
  157. 157 Llama 4 Maverick 890
  158. 158 Grok-3 mini 881
  159. 159 GPT-4o Mar 2025 876
  160. 160 Olmo 3.1 32B Think 850
  161. 161 Codestral 2 (2501) 843
  162. 162 Devstral Small 1.1 830
  163. 163 Mistral Large 2.1 (2411) 815
  164. 164 Llama 4 Scout 759

Source: Official DesignArena UI Component leaderboard

#EntryScoreSettingDateSource
1 GPT-6 Astra 1,401 xhigh Sep 2026 Official leaderboard
2 Kimi K3.1 Pro 1,388 Oct 2026 Official leaderboard
3 Claude Sonnet 5.5 1,386 Sep 2026 Official leaderboard
4 Claude Opus 5.5 1,379 Sep 2026 Official leaderboard
5 Muse Wonder 1,368 max Oct 2026 Official leaderboard
6 GPT-6.1 Sol 1,358 Sep 2026 Official leaderboard
7 Claude Fable 5.2 1,357 Nov 2026 Official leaderboard
8 Muse Spark 1.3 1,354 xhigh, max Sep 2026 Official leaderboard
9 Ling-3.1-flash 1,351 Oct 2026 Official leaderboard
10 Gemini 4 Carbon 1,349 Oct 2026 Official leaderboard
11 Claude Opus 5 1,341 Jul 2026 Official leaderboard
12 Kimi K3 1,339 Jul 2026 Official leaderboard
13 Qwen 3.8 Max 1,336 Aug 2026 Official leaderboard
14 Gemini 4 Argon 1,336 Sep 2026 Official leaderboard
15 Claude Fable 5.1 1,333 Sep 2026 Official leaderboard
16 MiMo-V2.6 Pro 1,333 Sep 2026 Official leaderboard
17 GPT-5.6 Sol 1,327 medium, xhigh Jul 2026 Official leaderboard
18 Muse Spark Open 1,314 xhigh Oct 2026 Official leaderboard
19 Muse Ember 1,312 Nov 2026 Official leaderboard
20 Claude Fable 5 1,311 Jun 2026 Official leaderboard
21 Gemini 3.8 Flash 1,306 Sep 2026 Official leaderboard
22 DeepSeek V4.1 Flash 1,304 Sep 2026 Official leaderboard
23 Muse Spark 1.2 1,301 Aug 2026 Official leaderboard
24 GLM-5.3-Flash 1,298 Aug 2026 Official leaderboard
25 MiniMax-M3.1-Flash 1,293 Oct 2026 Official leaderboard
26 GPT-6 Luna 1,292 Sep 2026 Official leaderboard
27 Gemini 3.6 Flash 1,291 Jul 2026 Official leaderboard
28 GPT-6.1 Terra 1,289 Oct 2026 Official leaderboard
29 GLM-5.2 1,288 Jun 2026 Official leaderboard
30 Gemini 3.7 Flash 1,288 Aug 2026 Official leaderboard
31 Grok 4.6 1,285 Aug 2026 Official leaderboard
32 GPT-6 Sol 1,285 Sep 2026 Official leaderboard
33 GLM-5.3 1,284 Aug 2026 Official leaderboard
34 Muse Spark 1.1 1,281 Jul 2026 Official leaderboard
35 Mistral Large 4 1,279 Oct 2026 Official leaderboard
36 Claude Opus 4.6 1,275 Feb 2026 Official leaderboard
37 Grok 4.7 1,274 Sep 2026 Official leaderboard
38 Claude Haiku 5.5 1,274 Oct 2026 Official leaderboard
39 Gemini 3 Pro 1,271 Nov 2025 Official leaderboard
40 Grok 4.5 1,270 Jul 2026 Official leaderboard
41 Claude Sonnet 5 1,269 Jun 2026 Official leaderboard
42 Gemini 3.5 Flash 1,266 May 2026 Official leaderboard
43 Qwen3.7-Max 1,264 May 2026 Official leaderboard
44 Claude Sonnet 4.6 1,263 Feb 2026 Official leaderboard
45 Deca 2.5 Mini 1,262 Official leaderboard
46 GLM-5.1 1,260 Apr 2026 Official leaderboard
47 Beam 1.0 1,259 max Oct 2026 Official leaderboard
48 Kimi K2.6 1,258 Apr 2026 Official leaderboard
49 Qwen3.7-Plus 1,256 Jun 2026 Official leaderboard
50 GLM-5-Turbo 1,254 Mar 2026 Official leaderboard
51 Kimi K2.7 Code 1,252 Jun 2026 Official leaderboard
52 Deca 2.6 Ultra 1,251 Official leaderboard
53 GPT-5.5 1,250 Apr 2026 Official leaderboard
54 MiMo-V2.5 1,250 Apr 2026 Official leaderboard
55 GPT-5.6 Terra 1,250 Jul 2026 Official leaderboard
56 Claude Opus 4.8 1,249 May 2026 Official leaderboard
57 MiMo-V2.5-Pro 1,244 Apr 2026 Official leaderboard
58 GPT-5.6 Luna 1,241 Jul 2026 Official leaderboard
59 MiniMax-M3 1,236 Jun 2026 Official leaderboard
60 Muse Spark 1,235 Apr 2026 Official leaderboard
61 GPT-5.4 1,231 medium Mar 2026 Official leaderboard
62 Deca 2.6 Pro 1,229 Official leaderboard
63 Claude Opus 4.5 1,229 Nov 2025 Official leaderboard
64 GPT-5.1-Codex 1,228 Nov 2025 Official leaderboard
65 Kimi K2.5 1,228 thinking Jan 2026 Official leaderboard
66 DeepSeek V4 Flash 0731 1,228 Jul 2026 Official leaderboard
67 Mercury 3 1,226 Oct 2026 Official leaderboard
68 Qwen 3.6 Plus 1,223 Mar 2026 Official leaderboard
69 GLM-5 1,222 Feb 2026 Official leaderboard
70 Gemini 3.1 Pro 1,216 Feb 2026 Official leaderboard
71 MiniMax-M2.1 1,214 Dec 2025 Official leaderboard
72 DeepSeek-V4-Pro 1,212 Apr 2026 Official leaderboard
73 Nex N2 Pro 1,211 Official leaderboard
74 Deca 2.5 Ultra 1,206 2 Official leaderboard
75 MiniMax-M2.7 1,203 Mar 2026 Official leaderboard
76 GLM-5V-Turbo 1,202 Apr 2026 Official leaderboard
77 glm-4.7-flash 1,199 Jan 2026 Official leaderboard
78 Muse Glimmer 30B 1,198 xhigh Official leaderboard
79 Tencent Hy3 1,195 Jul 2026 Official leaderboard
80 Tencent Hy4 1,194 Nov 2026 Official leaderboard
81 Grok 4.20 Beta 1,192 reasoning Official leaderboard
82 GLM-4.7 1,190 Dec 2025 Official leaderboard
83 Inkling 1,190 Jul 2026 Official leaderboard
84 Grok 4.3 1,186 Apr 2026 Official leaderboard
85 GPT-5.2 1,181 medium Dec 2025 Official leaderboard
86 DeepSeek-V3.1 1,173 thinking Aug 2025 Official leaderboard
87 GPT-5 1,171 minimal, high Aug 2025 Official leaderboard
88 Qwen3.5 Plus 02-15 1,169 Official leaderboard
89 Step 3.7 Flash 1,168 May 2026 Official leaderboard
90 MiniMax-M2.5 1,164 Feb 2026 Official leaderboard
91 Claude Sonnet 4.5 1,163 thinking Sep 2025 Official leaderboard
92 Flint 1,160 Official leaderboard
93 DeepSeek-V3.2-Exp 1,158 Sep 2025 Official leaderboard
94 GPT-5.1 1,157 medium Nov 2025 Official leaderboard
95 Claude Opus 4.1 1,156 thinking Aug 2025 Official leaderboard
96 v0-1.5-lg 1,155 Official leaderboard
97 DeepSeek-V4-Flash 1,155 Apr 2026 Official leaderboard
98 Qwen3.5 397B-A17B 1,154 Feb 2026 Official leaderboard
99 Nemotron 3.5 Super VL 1,154 Oct 2026 Official leaderboard
100 Claude Opus 4 1,144 May 2025 Official leaderboard
101 Claude 3.7 Sonnet 1,142 Feb 2025 Official leaderboard
102 Solar Pro 4 1,141 Aug 2026 Official leaderboard
103 AGI-01 Swift 1,140 Official leaderboard
104 DeepSeek-V3.2 1,140 Dec 2025 Official leaderboard
105 GLM-4.6 1,139 Sep 2025 Official leaderboard
106 GLM-4.5 1,135 Aug 2025 Official leaderboard
107 MiMo-V2-Flash 1,135 Dec 2025 Official leaderboard
108 Gemini 2.5 Pro Jun 2025 1,131 Jun 2025 Official leaderboard
109 MiniMax M2 Stable 1,128 Official leaderboard
110 GPT-5.3 Codex 1,128 Feb 2026 Official leaderboard
111 Nemotron 3 Ultra 1,128 Jun 2026 Official leaderboard
112 Claude Sonnet 4 1,120 May 2025 Official leaderboard
113 GLM-4.5-Air 1,114 Jul 2025 Official leaderboard
114 Qwen3 Coder 480B A35B Instruct 1,104 Official leaderboard
115 GPT-5 mini (Default) 1,097 Official leaderboard
116 DeepSeek-R1 May 2025 1,094 May 2025 Official leaderboard
117 Mistral Large 3 1,093 Dec 2025 Official leaderboard
118 Solar Mini 4 1,093 Oct 2026 Official leaderboard
119 Mistral Medium 3.1 1,091 Aug 2025 Official leaderboard
120 Claude Haiku 4.5 1,091 Oct 2025 Official leaderboard
121 DeepSeek-V3 Mar 2025 1,084 Mar 2025 Official leaderboard
122 Gemini 2.5 Flash Sep 2025 1,084 Sep 2025 Official leaderboard
123 AesCoder-4B 1,077 Official leaderboard
124 GPT-5.1-Codex-mini 1,072 Nov 2025 Official leaderboard
125 Mistral Medium 3.5 1,070 Apr 2026 Official leaderboard
126 Qwen3-Max 1,069 Sep 2025 Official leaderboard
127 Prime Intellect: INTELLECT-3 1,068 Official leaderboard
128 Grok 4 Fast 1,063 Sep 2025 Official leaderboard
129 Gemini 3.1 Flash-Lite 1,063 Mar 2026 Official leaderboard
130 GPT-5 nano (Default) 1,058 Official leaderboard
131 Grok 4.1 Fast 1,044 reasoning Nov 2025 Official leaderboard
132 Qwen3-Coder-30B-A3B-Instruct 1,037 Jul 2025 Official leaderboard
133 Trinity Large Thinking 1,036 Apr 2026 Official leaderboard
134 Grok 3 1,028 Apr 2025 Official leaderboard
135 Kimi K2 Sep 2025 1,021 Sep 2025 Official leaderboard
136 Mistral Medium 3 (2505) 1,016 Official leaderboard
137 Gemini 2.5 Flash-Lite Sep 2025 1,016 Sep 2025 Official leaderboard
138 Magistral Medium 1.2 (2509) 1,014 Official leaderboard
139 Codestral 2508 1,008 Official leaderboard
140 o3 1,008 Apr 2025 Official leaderboard
141 v0-1.5-md 1,001 Official leaderboard
142 GPT-4.1 991 Apr 2025 Official leaderboard
143 grok-code-fast-1 981 Aug 2025 Official leaderboard
144 Grok 4 977 Jul 2025 Official leaderboard
145 o4-mini 974 Apr 2025 Official leaderboard
146 Devstral Medium 970 Jul 2025 Official leaderboard
147 Mercury 2 970 Feb 2026 Official leaderboard
148 Qwen3-235B-A22B-Instruct-2507 956 Official leaderboard
149 GPT-4.1 mini 952 Apr 2025 Official leaderboard
150 Qwen3-235B-A22B 950 Apr 2025 Official leaderboard
151 Qwen3-30B-A3B 937 Apr 2025 Official leaderboard
152 Qwen3-235B-A22B-Thinking Jul 2025 936 Jul 2025 Official leaderboard
153 Magistral Small 1.2 (2509) 921 Official leaderboard
154 gpt-oss-120b 918 Aug 2025 Official leaderboard
155 GPT-4.1 nano 907 Apr 2025 Official leaderboard
156 Mistral Small 3.2 898 Jun 2025 Official leaderboard
157 Llama 4 Maverick 890 Apr 2025 Official leaderboard
158 Grok-3 mini 881 Apr 2025 Official leaderboard
159 GPT-4o Mar 2025 876 Mar 2025 Official leaderboard
160 Olmo 3.1 32B Think 850 Dec 2025 Official leaderboard
161 Codestral 2 (2501) 843 Official leaderboard
162 Devstral Small 1.1 830 Jul 2025 Official leaderboard
163 Mistral Large 2.1 (2411) 815 Official leaderboard
164 Llama 4 Scout 759 Apr 2025 Official leaderboard

More effort adds a median 15 elo

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
GPT-6 Astra
xhigh, max xhigh
1,401
Muse Spark 1.3
xhigh
1,354
GPT-5.6 Sol
medium
1,327
DeepSeek V4.1 Flash
1017
1,304
Claude Opus 4.6
thinking
1,275
GPT-5.4
none medium
1,231
Tencent Hy3
low
1,195
GPT-5.2
none
1,181
DeepSeek-V3.1
thinking
1,173
GPT-5
minimal minimal, hi…
1,171
Claude Sonnet 4.5
thinking
1,163
GPT-5.1
none
1,157

Source: Official DesignArena UI Component leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-6 Astra xhigh 1,401
Kimi K3.1 Pro 1,388
Claude Sonnet 5.5 1,386
Claude Opus 5.5 1,379
Muse Wonder max 1,368
GPT-6 Astra xhigh, max 1,366
GPT-6.1 Sol 1,358
Claude Fable 5.2 1,357
Muse Spark 1.3 xhigh, max 1,354
Ling-3.1-flash 1,351
Gemini 4 Carbon 1,349
Muse Spark 1.3 xhigh 1,348
Claude Opus 5 1,341
Kimi K3 1,339
Qwen 3.8 Max 1,336
Gemini 4 Argon 1,336
Claude Fable 5.1 1,333
MiMo-V2.6 Pro 1,333
GPT-5.6 Sol medium, xhigh 1,327
GPT-5.6 Sol medium 1,316
Muse Spark Open xhigh 1,314
Muse Ember 1,312
Claude Fable 5 1,311
Gemini 3.8 Flash 1,306
DeepSeek V4.1 Flash 1,304
Muse Spark 1.2 1,301
GLM-5.3-Flash 1,298
MiniMax-M3.1-Flash 1,293
GPT-6 Luna 1,292
Gemini 3.6 Flash 1,291
GPT-6.1 Terra 1,289
GLM-5.2 1,288
Gemini 3.7 Flash 1,288
DeepSeek V4.1 Flash 1017 1,286
Grok 4.6 1,285
GPT-6 Sol 1,285
GLM-5.3 1,284
Muse Spark 1.1 1,281
Mistral Large 4 1,279
Claude Opus 4.6 1,275
Grok 4.7 1,274
Claude Haiku 5.5 1,274
Claude Opus 4.6 thinking 1,272
Gemini 3 Pro 1,271
Grok 4.5 1,270
Claude Sonnet 5 1,269
Gemini 3.5 Flash 1,266
Qwen3.7-Max 1,264
Claude Sonnet 4.6 1,263
GLM-5.1 1,260
Beam 1.0 max 1,259
Kimi K2.6 1,258
Qwen3.7-Plus 1,256
GLM-5-Turbo 1,254
Kimi K2.7 Code 1,252
GPT-5.5 1,250
MiMo-V2.5 1,250
GPT-5.6 Terra 1,250
Claude Opus 4.8 1,249
MiMo-V2.5-Pro 1,244
GPT-5.6 Luna 1,241
MiniMax-M3 1,236
Muse Spark 1,235
GPT-5.4 medium 1,231
Claude Opus 4.5 1,229
GPT-5.1-Codex 1,228
Kimi K2.5 thinking 1,228
DeepSeek V4 Flash 0731 1,228
Mercury 3 1,226
Qwen 3.6 Plus 1,223
GLM-5 1,222
Gemini 3.1 Pro 1,216
MiniMax-M2.1 1,214
DeepSeek-V4-Pro 1,212
GPT-5.4 low 1,206
MiniMax-M2.7 1,203
GLM-5V-Turbo 1,202
glm-4.7-flash 1,199
Tencent Hy3 1,195
Tencent Hy4 1,194
GLM-4.7 1,190
GPT-5.4 none 1,190
Inkling 1,190
Grok 4.3 1,186
GPT-5.2 medium 1,181
DeepSeek-V3.1 thinking 1,173
GPT-5.2 high 1,173
GPT-5 minimal, high 1,171
Step 3.7 Flash 1,168
GPT-5.2 none 1,166
MiniMax-M2.5 1,164
Claude Sonnet 4.5 thinking 1,163
DeepSeek-V3.2-Exp 1,158
GPT-5.1 medium 1,157
Claude Opus 4.1 thinking 1,156
DeepSeek-V4-Flash 1,155
Qwen3.5 397B-A17B 1,154
Tencent Hy3 low 1,154
Nemotron 3.5 Super VL 1,154
Claude Sonnet 4.5 1,152
GPT-5.2 low 1,150
GPT-5.1 high 1,149
Claude Opus 4.1 1,147
GPT-5.1 low 1,145
Claude Opus 4 1,144
Claude 3.7 Sonnet 1,142
Solar Pro 4 1,141
DeepSeek-V3.2 1,140
GLM-4.6 1,139
GPT-5 minimal 1,136
GLM-4.5 1,135
MiMo-V2-Flash 1,135
GPT-5.1 none 1,134
Gemini 2.5 Pro Jun 2025 1,131
GPT-5.3 Codex 1,128
Nemotron 3 Ultra 1,128
Claude Sonnet 4 1,120
Claude Sonnet 4 thinking 1,116
GLM-4.5-Air 1,114
DeepSeek-R1 May 2025 1,094
Mistral Large 3 1,093
Solar Mini 4 1,093
Mistral Medium 3.1 1,091
Claude Haiku 4.5 1,091
DeepSeek-V3 Mar 2025 1,084
Gemini 2.5 Flash Sep 2025 1,084
DeepSeek-V3.1 1,076
GPT-5.1-Codex-mini 1,072
Mistral Medium 3.5 1,070
Qwen3-Max 1,069
Grok 4 Fast 1,063
Gemini 3.1 Flash-Lite 1,063
Grok 4.1 Fast reasoning 1,044
Qwen3-Coder-30B-A3B-Instruct 1,037
Trinity Large Thinking 1,036
Grok 3 1,028
Grok 4.1 Fast 1,028
Kimi K2 Sep 2025 1,021
Grok 4 Fast reasoning 1,019
Gemini 2.5 Flash-Lite Sep 2025 1,016
o3 1,008
Gemini 2.5 Flash Sep 2025 996
GPT-4.1 991
grok-code-fast-1 981
Grok 4 977
o4-mini 974
Devstral Medium 970
Mercury 2 970
GPT-4.1 mini 952
Qwen3-235B-A22B 950
Qwen3-30B-A3B 937
Qwen3-235B-A22B-Thinking Jul 2025 936
gpt-oss-120b 918
GPT-4.1 nano 907
Mistral Small 3.2 898
Llama 4 Maverick 890
Grok-3 mini 881
GPT-4o Mar 2025 876
Olmo 3.1 32B Think 850
Devstral Small 1.1 830
Llama 4 Scout 759

14 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.

Anthropic OpenAI DeepSeek Google DeepMind Moonshot AI Meta
Own leaderboard 14 leaders
GPT-6 Astra 1,401

Source: Official DesignArena UI Component 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
600 800 1,000 1,200 1,400 1,600 2025 2026 2027 600 800 1,000 1,200 1,400 1,600 2025 2026 2027

Source: Official DesignArena UI Component 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