Benchmark Almanac

DesignArena Code

GPT-6 Astra 1,396 leads since September 2026
Not in model cards

Users give a prompt, and anonymous models each build a website, UI component, browser game, data visualization or 3D scene as one HTML file. Users vote, and votes become Bradley-Terry ratings.

The Code overall board combines the Website, UI Component, Game Dev, Data Visualization and 3D boards. It listed 190 models on November 1st, 2026. Launched on Hacker News on July 12th, 2025. DesignArena runs 40 boards in all, from agentic web apps and slides to image, video and speech.

Released
July 2025
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
173
Entries on its own leaderboard, 143 models we track

GPT-6 Astra leads with 1,396

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.

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

  1. 1,396 GPT-6 Astra
  2. 1,394 Muse Wonder
  3. 1,385 Claude Opus 5.5
  4. 1,384 Inkling 2 Small
  5. 1,374 Claude Fable 5.2
  6. 1,361 GPT-6.1 Sol
  7. 1,354 Muse Spark 1.3
  8. 1,352 Kimi K3
  9. 1,351 Kimi K3.1 Pro
  10. 1,348 Muse Ember
  11. 1,347 Claude Sonnet 5.5
  12. 1,339 GPT-5.6 Sol
  13. 1,331 Claude Fable 5.1
  14. 1,331 Gemini 4 Carbon
  15. 1,327 Gemini 4 Argon
  16. 1,318 Claude Opus 5
  17. 1,316 MiMo-V2.6 Pro
  18. 1,310 DeepSeek V4.1 Flash
  19. 1,307 Muse Spark 1.2
  20. 1,304 Ling-3.1-flash
  21. 1,303 Muse Spark Open
  22. 1,302 Claude Fable 5
  23. 1,300 GPT-6.1 Terra
  24. 1,299 Gemini 3.7 Flash
  25. 1,297 MiniMax-M3.1-Flash
  26. 1,295 GPT-6 Sol
  27. 1,293 Gemini 3.8 Flash
  28. 1,288 Qwen 3.8 Max
  29. 1,285 GLM-5.3
  30. 1,281 Grok 4.6
  31. 1,280 Gemini 3.6 Flash
  32. 1,279 Claude Opus 4.6
  33. 1,278 GLM-5.2
  34. 1,278 Claude Haiku 5.5
  35. 1,270 Deca 2.6 Ultra
  36. 1,268 Claude Sonnet 5
  37. 1,268 GLM-5.3-Flash
  38. 1,267 Muse Spark 1.1
  39. 1,267 Grok 4.7
  40. 1,266 Claude Sonnet 4.6
  41. 1,266 Qwen3.7-Max
  42. 1,265 Grok 4.5
  43. 1,264 Deca 2.5 Mini
  44. 1,263 Kimi K2.6
  45. 1,262 MiMo-V2.5-Pro
  46. 1,262 GPT-6 Luna
  47. 1,261 GLM-5.1
  48. 1,260 Qwen3.7-Plus
  49. 1,257 Gemini 3.5 Flash
  50. 1,256 Deca 2.5 Ultra
  51. 1,256 Tencent Hy4
  52. 1,255 Mistral Large 4
  53. 1,253 GLM-5-Turbo
  54. 1,253 MiMo-V2.5
  55. 1,250 GPT-5.5
  56. 1,245 Deca 2.6 Pro
  57. 1,244 Claude Opus 4.8
  58. 1,244 Kimi K2.7 Code
  59. 1,242 Muse Spark
  60. 1,237 GLM-5
  61. 1,237 MiniMax-M3
  62. 1,234 Gemini 3 Pro
  63. 1,232 Claude Opus 4.5
  64. 1,231 DeepSeek-V4-Pro
  65. 1,230 GPT-5.6 Terra
  66. 1,227 Kimi K2.5
  67. 1,227 Qwen 3.6 Plus
  68. 1,227 Beam 1.0
  69. 1,225 Mercury 3
  70. 1,224 Nex N2 Pro
  71. 1,224 MiniMax-M2.7
  72. 1,221 Gemini 3.1 Pro
  73. 1,221 GPT-5.6 Luna
  74. 1,220 DeepSeek V4 Flash 0731
  75. 1,219 GLM-5V-Turbo
  76. 1,216 Tencent Hy3
  77. 1,211 Grok 4.20 Beta
  78. 1,207 GLM-4.7
  79. 1,202 GPT-5.4 (Design Skill, Medium)
  80. 1,200 GPT-5.4
  81. 1,197 MiniMax-M2.5
  82. 1,196 DeepSeek-V4-Flash
  83. 1,195 Inkling
  84. 1,181 MiniMax-M2.1
  85. 1,180 Gemini 3 Flash
  86. 1,177 Deca 2.6 Mini
  87. 1,176 Grok 4.3
  88. 1,173 Claude Sonnet 4.5
  89. 1,171 Solar Pro 4
  90. 1,170 Qwen3.5 397B-A17B
  91. 1,170 Step 3.7 Flash
  92. 1,168 glm-4.7-flash
  93. 1,167 Claude 3.7 Sonnet
  94. 1,166 DeepSeek-V3.1
  95. 1,161 Muse Glimmer 30B
  96. 1,161 Claude Opus 4.1
  97. 1,161 DeepSeek-V3.2-Exp
  98. 1,161 GPT-5.1
  99. 1,160 GPT-5.2
  100. 1,158 GPT-5
  101. 1,157 Qwen3.5 Plus 02-15
  102. 1,156 GLM-4.6
  103. 1,155 GLM-4.5
  104. 1,154 DeepSeek-V3.2
  105. 1,151 Claude Opus 4
  106. 1,146 MiMo-V2-Flash
  107. 1,143 Gemini 2.5 Pro
  108. 1,143 GPT-5.1-Codex
  109. 1,142 Flint
  110. 1,140 Nemotron 3.5 Super VL
  111. 1,135 GPT-5.3 Codex
  112. 1,134 v0-1.5-lg
  113. 1,133 Qwen3 Coder 480B A35B Instruct
  114. 1,133 Nemotron 3 Ultra
  115. 1,132 Claude Sonnet 4
  116. 1,132 Mistral Large 3
  117. 1,129 DeepSeek-R1
  118. 1,128 GLM-4.5-Air
  119. 1,127 Solar Mini 4
  120. 1,126 AGI-01 Swift
  121. 1,125 MiniMax M2 Stable
  122. 1,115 AesCoder-4B
  123. 1,112 Mistral Medium 3.5
  124. 1,111 Mistral Medium 3.1
  125. 1,108 Trinity Large Thinking
  126. 1,105 GPT-5 mini (Default)
  127. 1,105 Claude Haiku 4.5
  128. 1,100 Qwen3-Max
  129. 1,098 DeepSeek-V3
  130. 1,097 Prime Intellect: INTELLECT-3
  131. 1,094 Gemini 2.5 Flash
  132. 1,091 Grok 4 Fast
  133. 1,088 Kimi K2 0905 Preview
  134. 1,084 GPT-5.1-Codex-mini
  135. 1,082 Grok 4.1 Fast
  136. 1,075 GPT-5 nano (Default)
  137. 1,074 Kimi K2 Turbo Preview
  138. 1,071 Gemini 2.5 Flash-Lite
  139. 1,063 Gemini 3.1 Flash-Lite
  140. 1,059 Mistral Medium 3 (2505)
  141. 1,055 Ministral 3 14B
  142. 1,047 v0-1.5-md
  143. 1,043 Grok 3
  144. 1,043 Ministral 3 8B Instruct
  145. 1,029 Qwen3-235B-A22B-Instruct-2507
  146. 1,024 Kimi K2
  147. 1,023 Magistral Medium 1.2 (2509)
  148. 1,023 Qwen3-235B-A22B-Thinking
  149. 1,016 GPT-4.1
  150. 1,010 o3
  151. 1,006 Grok 4
  152. 1,003 Devstral Medium
  153. 1,000 Ministral 3 3B Instruct
  154. 997 Codestral 2508
  155. 997 Mercury 2
  156. 992 Qwen3-235B-A22B
  157. 989 grok-code-fast-1
  158. 984 GPT-4.1 mini
  159. 976 Magistral Small 1.2 (2509)
  160. 966 o4-mini
  161. 965 Olmo 3.1 32B Think
  162. 953 GPT-4.1 nano
  163. 953 gpt-oss-120b
  164. 932 Qwen3-30B-A3B
  165. 921 Grok-3 mini
  166. 919 Llama 3.1 Nemotron Ultra 253B
  167. 897 Mistral Small 3.2
  168. 870 Llama 4 Maverick
  169. 853 Mistral Large 2.1 (2411)
  170. 851 GPT-4o
  171. 824 Codestral 2 (2501)
  172. 798 Devstral Small 1.1
  173. 780 Llama 4 Scout

Bars start at 600.

  1. 1 GPT-6 Astra xhigh 1,396
  2. 2 Muse Wonder max 1,394
  3. 3 Claude Opus 5.5 1,385
  4. 4 Inkling 2 Small 1,384
  5. 5 Claude Fable 5.2 1,374
  6. 6 GPT-6.1 Sol 1,361
  7. 7 Muse Spark 1.3 xhigh, max 1,354
  8. 8 Kimi K3 1,352
  9. 9 Kimi K3.1 Pro 1,351
  10. 10 Muse Ember 1,348
  11. 11 Claude Sonnet 5.5 1,347
  12. 12 GPT-5.6 Sol medium, xhigh 1,339
  13. 13 Claude Fable 5.1 1,331
  14. 14 Gemini 4 Carbon 1,331
  15. 15 Gemini 4 Argon 1,327
  16. 16 Claude Opus 5 1,318
  17. 17 MiMo-V2.6 Pro 1,316
  18. 18 DeepSeek V4.1 Flash 1,310
  19. 19 Muse Spark 1.2 1,307
  20. 20 Ling-3.1-flash 1,304
  21. 21 Muse Spark Open xhigh 1,303
  22. 22 Claude Fable 5 1,302
  23. 23 GPT-6.1 Terra 1,300
  24. 24 Gemini 3.7 Flash 1,299
  25. 25 MiniMax-M3.1-Flash 1,297
  26. 26 GPT-6 Sol 1,295
  27. 27 Gemini 3.8 Flash 1,293
  28. 28 Qwen 3.8 Max 1,288
  29. 29 GLM-5.3 1,285
  30. 30 Grok 4.6 1,281
  31. 31 Gemini 3.6 Flash 1,280
  32. 32 Claude Opus 4.6 1,279
  33. 33 GLM-5.2 1,278
  34. 34 Claude Haiku 5.5 1,278
  35. 35 Deca 2.6 Ultra 1,270
  36. 36 Claude Sonnet 5 1,268
  37. 37 GLM-5.3-Flash 1,268
  38. 38 Muse Spark 1.1 1,267
  39. 39 Grok 4.7 1,267
  40. 40 Claude Sonnet 4.6 1,266
  41. 41 Qwen3.7-Max 1,266
  42. 42 Grok 4.5 1,265
  43. 43 Deca 2.5 Mini 1,264
  44. 44 Kimi K2.6 1,263
  45. 45 MiMo-V2.5-Pro 1,262
  46. 46 GPT-6 Luna 1,262
  47. 47 GLM-5.1 1,261
  48. 48 Qwen3.7-Plus 1,260
  49. 49 Gemini 3.5 Flash 1,257
  50. 50 Deca 2.5 Ultra 2 1,256
  51. 51 Tencent Hy4 1,256
  52. 52 Mistral Large 4 1,255
  53. 53 GLM-5-Turbo 1,253
  54. 54 MiMo-V2.5 1,253
  55. 55 GPT-5.5 1,250
  56. 56 Deca 2.6 Pro 1,245
  57. 57 Claude Opus 4.8 1,244
  58. 58 Kimi K2.7 Code 1,244
  59. 59 Muse Spark 1,242
  60. 60 GLM-5 1,237
  61. 61 MiniMax-M3 1,237
  62. 62 Gemini 3 Pro 1,234
  63. 63 Claude Opus 4.5 1,232
  64. 64 DeepSeek-V4-Pro 1,231
  65. 65 GPT-5.6 Terra 1,230
  66. 66 Kimi K2.5 thinking 1,227
  67. 67 Qwen 3.6 Plus 1,227
  68. 68 Beam 1.0 max 1,227
  69. 69 Mercury 3 1,225
  70. 70 Nex N2 Pro 1,224
  71. 71 MiniMax-M2.7 1,224
  72. 72 Gemini 3.1 Pro 1,221
  73. 73 GPT-5.6 Luna 1,221
  74. 74 DeepSeek V4 Flash 0731 1,220
  75. 75 GLM-5V-Turbo 1,219
  76. 76 Tencent Hy3 1,216
  77. 77 Grok 4.20 Beta reasoning 1,211
  78. 78 GLM-4.7 1,207
  79. 79 GPT-5.4 (Design Skill, Medium) 1,202
  80. 80 GPT-5.4 medium 1,200
  81. 81 MiniMax-M2.5 1,197
  82. 82 DeepSeek-V4-Flash 1,196
  83. 83 Inkling 1,195
  84. 84 MiniMax-M2.1 1,181
  85. 85 Gemini 3 Flash 1,180
  86. 86 Deca 2.6 Mini 1,177
  87. 87 Grok 4.3 1,176
  88. 88 Claude Sonnet 4.5 1,173
  89. 89 Solar Pro 4 1,171
  90. 90 Qwen3.5 397B-A17B 1,170
  91. 91 Step 3.7 Flash 1,170
  92. 92 glm-4.7-flash 1,168
  93. 93 Claude 3.7 Sonnet 1,167
  94. 94 DeepSeek-V3.1 thinking 1,166
  95. 95 Muse Glimmer 30B xhigh 1,161
  96. 96 Claude Opus 4.1 thinking 1,161
  97. 97 DeepSeek-V3.2-Exp 1,161
  98. 98 GPT-5.1 high 1,161
  99. 99 GPT-5.2 medium 1,160
  100. 100 GPT-5 minimal, high 1,158
  101. 101 Qwen3.5 Plus 02-15 1,157
  102. 102 GLM-4.6 1,156
  103. 103 GLM-4.5 1,155
  104. 104 DeepSeek-V3.2 1,154
  105. 105 Claude Opus 4 1,151
  106. 106 MiMo-V2-Flash 1,146
  107. 107 Gemini 2.5 Pro Jun 2025 1,143
  108. 108 GPT-5.1-Codex 1,143
  109. 109 Flint 1,142
  110. 110 Nemotron 3.5 Super VL 1,140
  111. 111 GPT-5.3 Codex 1,135
  112. 112 v0-1.5-lg 1,134
  113. 113 Qwen3 Coder 480B A35B Instruct 1,133
  114. 114 Nemotron 3 Ultra 1,133
  115. 115 Claude Sonnet 4 1,132
  116. 116 Mistral Large 3 1,132
  117. 117 DeepSeek-R1 May 2025 1,129
  118. 118 GLM-4.5-Air 1,128
  119. 119 Solar Mini 4 1,127
  120. 120 AGI-01 Swift 1,126
  121. 121 MiniMax M2 Stable 1,125
  122. 122 AesCoder-4B 1,115
  123. 123 Mistral Medium 3.5 1,112
  124. 124 Mistral Medium 3.1 1,111
  125. 125 Trinity Large Thinking 1,108
  126. 126 GPT-5 mini (Default) 1,105
  127. 127 Claude Haiku 4.5 1,105
  128. 128 Qwen3-Max 1,100
  129. 129 DeepSeek-V3 Mar 2025 1,098
  130. 130 Prime Intellect: INTELLECT-3 1,097
  131. 131 Gemini 2.5 Flash Sep 2025 1,094
  132. 132 Grok 4 Fast 1,091
  133. 133 Kimi K2 0905 Preview 1,088
  134. 134 GPT-5.1-Codex-mini 1,084
  135. 135 Grok 4.1 Fast 1,082
  136. 136 GPT-5 nano (Default) 1,075
  137. 137 Kimi K2 Turbo Preview 1,074
  138. 138 Gemini 2.5 Flash-Lite Sep 2025 1,071
  139. 139 Gemini 3.1 Flash-Lite 1,063
  140. 140 Mistral Medium 3 (2505) 1,059
  141. 141 Ministral 3 14B 1,055
  142. 142 v0-1.5-md 1,047
  143. 143 Grok 3 1,043
  144. 144 Ministral 3 8B Instruct 1,043
  145. 145 Qwen3-235B-A22B-Instruct-2507 1,029
  146. 146 Kimi K2 Sep 2025 1,024
  147. 147 Magistral Medium 1.2 (2509) 1,023
  148. 148 Qwen3-235B-A22B-Thinking Jul 2025 1,023
  149. 149 GPT-4.1 1,016
  150. 150 o3 1,010
  151. 151 Grok 4 1,006
  152. 152 Devstral Medium 1,003
  153. 153 Ministral 3 3B Instruct 1,000
  154. 154 Codestral 2508 997
  155. 155 Mercury 2 997
  156. 156 Qwen3-235B-A22B 992
  157. 157 grok-code-fast-1 989
  158. 158 GPT-4.1 mini 984
  159. 159 Magistral Small 1.2 (2509) 976
  160. 160 o4-mini 966
  161. 161 Olmo 3.1 32B Think 965
  162. 162 GPT-4.1 nano 953
  163. 163 gpt-oss-120b 953
  164. 164 Qwen3-30B-A3B 932
  165. 165 Grok-3 mini 921
  166. 166 Llama 3.1 Nemotron Ultra 253B 919
  167. 167 Mistral Small 3.2 897
  168. 168 Llama 4 Maverick 870
  169. 169 Mistral Large 2.1 (2411) 853
  170. 170 GPT-4o Mar 2025 851
  171. 171 Codestral 2 (2501) 824
  172. 172 Devstral Small 1.1 798
  173. 173 Llama 4 Scout 780

Source: Official DesignArena Code leaderboard

#EntryScoreSettingDateSource
1 GPT-6 Astra 1,396 xhigh Sep 2026 Official leaderboard
2 Muse Wonder 1,394 max Oct 2026 Official leaderboard
3 Claude Opus 5.5 1,385 Sep 2026 Official leaderboard
4 Inkling 2 Small 1,384 Nov 2026 Official leaderboard
5 Claude Fable 5.2 1,374 Nov 2026 Official leaderboard
6 GPT-6.1 Sol 1,361 Sep 2026 Official leaderboard
7 Muse Spark 1.3 1,354 xhigh, max Sep 2026 Official leaderboard
8 Kimi K3 1,352 Jul 2026 Official leaderboard
9 Kimi K3.1 Pro 1,351 Oct 2026 Official leaderboard
10 Muse Ember 1,348 Nov 2026 Official leaderboard
11 Claude Sonnet 5.5 1,347 Sep 2026 Official leaderboard
12 GPT-5.6 Sol 1,339 medium, xhigh Jul 2026 Official leaderboard
13 Claude Fable 5.1 1,331 Sep 2026 Official leaderboard
14 Gemini 4 Carbon 1,331 Oct 2026 Official leaderboard
15 Gemini 4 Argon 1,327 Sep 2026 Official leaderboard
16 Claude Opus 5 1,318 Jul 2026 Official leaderboard
17 MiMo-V2.6 Pro 1,316 Sep 2026 Official leaderboard
18 DeepSeek V4.1 Flash 1,310 Sep 2026 Official leaderboard
19 Muse Spark 1.2 1,307 Aug 2026 Official leaderboard
20 Ling-3.1-flash 1,304 Oct 2026 Official leaderboard
21 Muse Spark Open 1,303 xhigh Oct 2026 Official leaderboard
22 Claude Fable 5 1,302 Jun 2026 Official leaderboard
23 GPT-6.1 Terra 1,300 Oct 2026 Official leaderboard
24 Gemini 3.7 Flash 1,299 Aug 2026 Official leaderboard
25 MiniMax-M3.1-Flash 1,297 Oct 2026 Official leaderboard
26 GPT-6 Sol 1,295 Sep 2026 Official leaderboard
27 Gemini 3.8 Flash 1,293 Sep 2026 Official leaderboard
28 Qwen 3.8 Max 1,288 Aug 2026 Official leaderboard
29 GLM-5.3 1,285 Aug 2026 Official leaderboard
30 Grok 4.6 1,281 Aug 2026 Official leaderboard
31 Gemini 3.6 Flash 1,280 Jul 2026 Official leaderboard
32 Claude Opus 4.6 1,279 Feb 2026 Official leaderboard
33 GLM-5.2 1,278 Jun 2026 Official leaderboard
34 Claude Haiku 5.5 1,278 Oct 2026 Official leaderboard
35 Deca 2.6 Ultra 1,270 Official leaderboard
36 Claude Sonnet 5 1,268 Jun 2026 Official leaderboard
37 GLM-5.3-Flash 1,268 Aug 2026 Official leaderboard
38 Muse Spark 1.1 1,267 Jul 2026 Official leaderboard
39 Grok 4.7 1,267 Sep 2026 Official leaderboard
40 Claude Sonnet 4.6 1,266 Feb 2026 Official leaderboard
41 Qwen3.7-Max 1,266 May 2026 Official leaderboard
42 Grok 4.5 1,265 Jul 2026 Official leaderboard
43 Deca 2.5 Mini 1,264 Official leaderboard
44 Kimi K2.6 1,263 Apr 2026 Official leaderboard
45 MiMo-V2.5-Pro 1,262 Apr 2026 Official leaderboard
46 GPT-6 Luna 1,262 Sep 2026 Official leaderboard
47 GLM-5.1 1,261 Apr 2026 Official leaderboard
48 Qwen3.7-Plus 1,260 Jun 2026 Official leaderboard
49 Gemini 3.5 Flash 1,257 May 2026 Official leaderboard
50 Deca 2.5 Ultra 1,256 2 Official leaderboard
51 Tencent Hy4 1,256 Nov 2026 Official leaderboard
52 Mistral Large 4 1,255 Oct 2026 Official leaderboard
53 GLM-5-Turbo 1,253 Mar 2026 Official leaderboard
54 MiMo-V2.5 1,253 Apr 2026 Official leaderboard
55 GPT-5.5 1,250 Apr 2026 Official leaderboard
56 Deca 2.6 Pro 1,245 Official leaderboard
57 Claude Opus 4.8 1,244 May 2026 Official leaderboard
58 Kimi K2.7 Code 1,244 Jun 2026 Official leaderboard
59 Muse Spark 1,242 Apr 2026 Official leaderboard
60 GLM-5 1,237 Feb 2026 Official leaderboard
61 MiniMax-M3 1,237 Jun 2026 Official leaderboard
62 Gemini 3 Pro 1,234 Nov 2025 Official leaderboard
63 Claude Opus 4.5 1,232 Nov 2025 Official leaderboard
64 DeepSeek-V4-Pro 1,231 Apr 2026 Official leaderboard
65 GPT-5.6 Terra 1,230 Jul 2026 Official leaderboard
66 Kimi K2.5 1,227 thinking Jan 2026 Official leaderboard
67 Qwen 3.6 Plus 1,227 Mar 2026 Official leaderboard
68 Beam 1.0 1,227 max Oct 2026 Official leaderboard
69 Mercury 3 1,225 Oct 2026 Official leaderboard
70 Nex N2 Pro 1,224 Official leaderboard
71 MiniMax-M2.7 1,224 Mar 2026 Official leaderboard
72 Gemini 3.1 Pro 1,221 Feb 2026 Official leaderboard
73 GPT-5.6 Luna 1,221 Jul 2026 Official leaderboard
74 DeepSeek V4 Flash 0731 1,220 Jul 2026 Official leaderboard
75 GLM-5V-Turbo 1,219 Apr 2026 Official leaderboard
76 Tencent Hy3 1,216 Jul 2026 Official leaderboard
77 Grok 4.20 Beta 1,211 reasoning Official leaderboard
78 GLM-4.7 1,207 Dec 2025 Official leaderboard
79 GPT-5.4 (Design Skill, Medium) 1,202 Official leaderboard
80 GPT-5.4 1,200 medium Mar 2026 Official leaderboard
81 MiniMax-M2.5 1,197 Feb 2026 Official leaderboard
82 DeepSeek-V4-Flash 1,196 Apr 2026 Official leaderboard
83 Inkling 1,195 Jul 2026 Official leaderboard
84 MiniMax-M2.1 1,181 Dec 2025 Official leaderboard
85 Gemini 3 Flash 1,180 Dec 2025 Official leaderboard
86 Deca 2.6 Mini 1,177 Official leaderboard
87 Grok 4.3 1,176 Apr 2026 Official leaderboard
88 Claude Sonnet 4.5 1,173 Sep 2025 Official leaderboard
89 Solar Pro 4 1,171 Aug 2026 Official leaderboard
90 Qwen3.5 397B-A17B 1,170 Feb 2026 Official leaderboard
91 Step 3.7 Flash 1,170 May 2026 Official leaderboard
92 glm-4.7-flash 1,168 Jan 2026 Official leaderboard
93 Claude 3.7 Sonnet 1,167 Feb 2025 Official leaderboard
94 DeepSeek-V3.1 1,166 thinking Aug 2025 Official leaderboard
95 Muse Glimmer 30B 1,161 xhigh Official leaderboard
96 Claude Opus 4.1 1,161 thinking Aug 2025 Official leaderboard
97 DeepSeek-V3.2-Exp 1,161 Sep 2025 Official leaderboard
98 GPT-5.1 1,161 high Nov 2025 Official leaderboard
99 GPT-5.2 1,160 medium Dec 2025 Official leaderboard
100 GPT-5 1,158 minimal, high Aug 2025 Official leaderboard
101 Qwen3.5 Plus 02-15 1,157 Official leaderboard
102 GLM-4.6 1,156 Sep 2025 Official leaderboard
103 GLM-4.5 1,155 Aug 2025 Official leaderboard
104 DeepSeek-V3.2 1,154 Dec 2025 Official leaderboard
105 Claude Opus 4 1,151 May 2025 Official leaderboard
106 MiMo-V2-Flash 1,146 Dec 2025 Official leaderboard
107 Gemini 2.5 Pro Jun 2025 1,143 Jun 2025 Official leaderboard
108 GPT-5.1-Codex 1,143 Nov 2025 Official leaderboard
109 Flint 1,142 Official leaderboard
110 Nemotron 3.5 Super VL 1,140 Oct 2026 Official leaderboard
111 GPT-5.3 Codex 1,135 Feb 2026 Official leaderboard
112 v0-1.5-lg 1,134 Official leaderboard
113 Qwen3 Coder 480B A35B Instruct 1,133 Official leaderboard
114 Nemotron 3 Ultra 1,133 Jun 2026 Official leaderboard
115 Claude Sonnet 4 1,132 May 2025 Official leaderboard
116 Mistral Large 3 1,132 Dec 2025 Official leaderboard
117 DeepSeek-R1 May 2025 1,129 May 2025 Official leaderboard
118 GLM-4.5-Air 1,128 Jul 2025 Official leaderboard
119 Solar Mini 4 1,127 Oct 2026 Official leaderboard
120 AGI-01 Swift 1,126 Official leaderboard
121 MiniMax M2 Stable 1,125 Official leaderboard
122 AesCoder-4B 1,115 Official leaderboard
123 Mistral Medium 3.5 1,112 Apr 2026 Official leaderboard
124 Mistral Medium 3.1 1,111 Aug 2025 Official leaderboard
125 Trinity Large Thinking 1,108 Apr 2026 Official leaderboard
126 GPT-5 mini (Default) 1,105 Official leaderboard
127 Claude Haiku 4.5 1,105 Oct 2025 Official leaderboard
128 Qwen3-Max 1,100 Sep 2025 Official leaderboard
129 DeepSeek-V3 Mar 2025 1,098 Mar 2025 Official leaderboard
130 Prime Intellect: INTELLECT-3 1,097 Official leaderboard
131 Gemini 2.5 Flash Sep 2025 1,094 Sep 2025 Official leaderboard
132 Grok 4 Fast 1,091 Sep 2025 Official leaderboard
133 Kimi K2 0905 Preview 1,088 Official leaderboard
134 GPT-5.1-Codex-mini 1,084 Nov 2025 Official leaderboard
135 Grok 4.1 Fast 1,082 Nov 2025 Official leaderboard
136 GPT-5 nano (Default) 1,075 Official leaderboard
137 Kimi K2 Turbo Preview 1,074 Official leaderboard
138 Gemini 2.5 Flash-Lite Sep 2025 1,071 Sep 2025 Official leaderboard
139 Gemini 3.1 Flash-Lite 1,063 Mar 2026 Official leaderboard
140 Mistral Medium 3 (2505) 1,059 Official leaderboard
141 Ministral 3 14B 1,055 Dec 2025 Official leaderboard
142 v0-1.5-md 1,047 Official leaderboard
143 Grok 3 1,043 Apr 2025 Official leaderboard
144 Ministral 3 8B Instruct 1,043 Dec 2025 Official leaderboard
145 Qwen3-235B-A22B-Instruct-2507 1,029 Official leaderboard
146 Kimi K2 Sep 2025 1,024 Sep 2025 Official leaderboard
147 Magistral Medium 1.2 (2509) 1,023 Official leaderboard
148 Qwen3-235B-A22B-Thinking Jul 2025 1,023 Jul 2025 Official leaderboard
149 GPT-4.1 1,016 Apr 2025 Official leaderboard
150 o3 1,010 Apr 2025 Official leaderboard
151 Grok 4 1,006 Jul 2025 Official leaderboard
152 Devstral Medium 1,003 Jul 2025 Official leaderboard
153 Ministral 3 3B Instruct 1,000 Dec 2025 Official leaderboard
154 Codestral 2508 997 Official leaderboard
155 Mercury 2 997 Feb 2026 Official leaderboard
156 Qwen3-235B-A22B 992 Apr 2025 Official leaderboard
157 grok-code-fast-1 989 Aug 2025 Official leaderboard
158 GPT-4.1 mini 984 Apr 2025 Official leaderboard
159 Magistral Small 1.2 (2509) 976 Official leaderboard
160 o4-mini 966 Apr 2025 Official leaderboard
161 Olmo 3.1 32B Think 965 Dec 2025 Official leaderboard
162 GPT-4.1 nano 953 Apr 2025 Official leaderboard
163 gpt-oss-120b 953 Aug 2025 Official leaderboard
164 Qwen3-30B-A3B 932 Apr 2025 Official leaderboard
165 Grok-3 mini 921 Apr 2025 Official leaderboard
166 Llama 3.1 Nemotron Ultra 253B 919 Official leaderboard
167 Mistral Small 3.2 897 Jun 2025 Official leaderboard
168 Llama 4 Maverick 870 Apr 2025 Official leaderboard
169 Mistral Large 2.1 (2411) 853 Official leaderboard
170 GPT-4o Mar 2025 851 Mar 2025 Official leaderboard
171 Codestral 2 (2501) 824 Official leaderboard
172 Devstral Small 1.1 798 Jul 2025 Official leaderboard
173 Llama 4 Scout 780 Apr 2025 Official leaderboard

More effort adds a median 19 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
1,396
Muse Spark 1.3
xhigh
1,354
GPT-5.6 Sol
medium medium, xhi…
1,339
DeepSeek V4.1 Flash
1017
1,310
Claude Opus 4.6
thinking
1,279
Tencent Hy3
low
1,216
GPT-5.4
none medium
1,200
Claude Sonnet 4.5
thinking
1,173
DeepSeek-V3.1
thinking
1,166
Claude Opus 4.1
thinking
1,161
GPT-5.1
none
1,161
GPT-5.2
none
1,160

Source: Official DesignArena Code leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-6 Astra xhigh 1,396
Muse Wonder max 1,394
Claude Opus 5.5 1,385
Inkling 2 Small 1,384
GPT-6 Astra xhigh, max 1,377
Claude Fable 5.2 1,374
GPT-6.1 Sol 1,361
Muse Spark 1.3 xhigh, max 1,354
Kimi K3 1,352
Muse Spark 1.3 xhigh 1,352
Kimi K3.1 Pro 1,351
Muse Ember 1,348
Claude Sonnet 5.5 1,347
GPT-5.6 Sol medium, xhigh 1,339
Claude Fable 5.1 1,331
Gemini 4 Carbon 1,331
Gemini 4 Argon 1,327
Claude Opus 5 1,318
MiMo-V2.6 Pro 1,316
DeepSeek V4.1 Flash 1,310
GPT-5.6 Sol medium 1,308
Muse Spark 1.2 1,307
Ling-3.1-flash 1,304
Muse Spark Open xhigh 1,303
Claude Fable 5 1,302
GPT-6.1 Terra 1,300
Gemini 3.7 Flash 1,299
MiniMax-M3.1-Flash 1,297
GPT-6 Sol 1,295
Gemini 3.8 Flash 1,293
Qwen 3.8 Max 1,288
DeepSeek V4.1 Flash 1017 1,287
GLM-5.3 1,285
Grok 4.6 1,281
Gemini 3.6 Flash 1,280
Claude Opus 4.6 1,279
GLM-5.2 1,278
Claude Haiku 5.5 1,278
Claude Opus 4.6 thinking 1,276
Claude Sonnet 5 1,268
GLM-5.3-Flash 1,268
Muse Spark 1.1 1,267
Grok 4.7 1,267
Claude Sonnet 4.6 1,266
Qwen3.7-Max 1,266
Grok 4.5 1,265
Kimi K2.6 1,263
MiMo-V2.5-Pro 1,262
GPT-6 Luna 1,262
GLM-5.1 1,261
Qwen3.7-Plus 1,260
Gemini 3.5 Flash 1,257
Tencent Hy4 1,256
Mistral Large 4 1,255
GLM-5-Turbo 1,253
MiMo-V2.5 1,253
GPT-5.5 1,250
Claude Opus 4.8 1,244
Kimi K2.7 Code 1,244
Muse Spark 1,242
GLM-5 1,237
MiniMax-M3 1,237
Gemini 3 Pro 1,234
Claude Opus 4.5 1,232
DeepSeek-V4-Pro 1,231
GPT-5.6 Terra 1,230
Kimi K2.5 thinking 1,227
Qwen 3.6 Plus 1,227
Beam 1.0 max 1,227
Mercury 3 1,225
MiniMax-M2.7 1,224
Gemini 3.1 Pro 1,221
GPT-5.6 Luna 1,221
DeepSeek V4 Flash 0731 1,220
GLM-5V-Turbo 1,219
Tencent Hy3 1,216
GLM-4.7 1,207
GPT-5.4 medium 1,200
MiniMax-M2.5 1,197
DeepSeek-V4-Flash 1,196
Inkling 1,195
MiniMax-M2.1 1,181
Gemini 3 Flash 1,180
Grok 4.3 1,176
Claude Sonnet 4.5 1,173
Claude Sonnet 4.5 thinking 1,172
Solar Pro 4 1,171
Qwen3.5 397B-A17B 1,170
GPT-5.4 low 1,170
Step 3.7 Flash 1,170
GPT-5.4 none 1,169
glm-4.7-flash 1,168
Claude 3.7 Sonnet 1,167
Tencent Hy3 low 1,167
DeepSeek-V3.1 thinking 1,166
Claude Opus 4.1 thinking 1,161
DeepSeek-V3.2-Exp 1,161
GPT-5.1 high 1,161
GPT-5.2 medium 1,160
GPT-5.2 none 1,159
GPT-5 minimal, high 1,158
GLM-4.6 1,156
GPT-5.2 low 1,156
GLM-4.5 1,155
Claude Opus 4.1 1,155
GPT-5 minimal 1,155
DeepSeek-V3.2 1,154
GPT-5.1 medium 1,152
Claude Opus 4 1,151
GPT-5.1 low 1,146
MiMo-V2-Flash 1,146
Gemini 2.5 Pro Jun 2025 1,143
GPT-5.1-Codex 1,143
GPT-5.1 none 1,141
GPT-5.2 high 1,141
Nemotron 3.5 Super VL 1,140
GPT-5.3 Codex 1,135
Nemotron 3 Ultra 1,133
Claude Sonnet 4 1,132
Mistral Large 3 1,132
DeepSeek-R1 May 2025 1,129
GLM-4.5-Air 1,128
Claude Sonnet 4 thinking 1,127
Solar Mini 4 1,127
Mistral Medium 3.5 1,112
Mistral Medium 3.1 1,111
Trinity Large Thinking 1,108
Claude Haiku 4.5 1,105
DeepSeek-V3.1 1,102
Qwen3-Max 1,100
DeepSeek-V3 Mar 2025 1,098
Gemini 2.5 Flash Sep 2025 1,094
Grok 4 Fast 1,091
GPT-5.1-Codex-mini 1,084
Grok 4.1 Fast 1,082
Grok 4.1 Fast reasoning 1,077
Gemini 2.5 Flash-Lite Sep 2025 1,071
Gemini 3.1 Flash-Lite 1,063
Ministral 3 14B 1,055
Gemini 2.5 Flash Sep 2025 1,050
Grok 3 1,043
Ministral 3 8B Instruct 1,043
Grok 4 Fast reasoning 1,041
Kimi K2 Sep 2025 1,024
Qwen3-235B-A22B-Thinking Jul 2025 1,023
GPT-4.1 1,016
o3 1,010
Grok 4 1,006
Devstral Medium 1,003
Ministral 3 3B Instruct 1,000
Mercury 2 997
Qwen3-235B-A22B 992
grok-code-fast-1 989
GPT-4.1 mini 984
o4-mini 966
Olmo 3.1 32B Think 965
GPT-4.1 nano 953
gpt-oss-120b 953
Qwen3-30B-A3B 932
Grok-3 mini 921
Mistral Small 3.2 897
Llama 4 Maverick 870
GPT-4o Mar 2025 851
Devstral Small 1.1 798
Llama 4 Scout 780

9 models have held number one on the official leaderboard since its release in July 2025

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

Anthropic Google DeepMind OpenAI Moonshot AI Meta
Own leaderboard 9 leaders
GPT-6 Astra 1,396

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

Source: Official DesignArena Code 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