Benchmark Almanac

DesignArena Data Visualization

Claude Opus 5.5 1,388 leads since September 2026
Not in model cards

Models turn a user prompt and its data into an interactive visualization as a single HTML file. Users compare the rendered charts and vote.

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
Data science and SQL
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
November 2026
Scores refreshed
November 12th, 2026
133
Entries on its own leaderboard, 118 models we track

Claude Opus 5.5 leads with 1,388

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.

133 entries. Scroll sideways for the rest. Bars start at 800.

  1. 1,388 Claude Opus 5.5
  2. 1,360 Claude Sonnet 5.5
  3. 1,351 Kimi K3
  4. 1,346 Claude Fable 5.1
  5. 1,343 Muse Spark 1.3
  6. 1,337 GPT-6 Astra
  7. 1,337 MiMo-V2.6 Pro
  8. 1,330 Muse Spark 1.2
  9. 1,327 Claude Opus 5
  10. 1,325 Gemini 4 Argon
  11. 1,322 GPT-5.6 Sol
  12. 1,317 Muse Ember
  13. 1,310 Gemini 3.7 Flash
  14. 1,304 Claude Fable 5
  15. 1,304 Kimi K3.1 Pro
  16. 1,290 Muse Wonder
  17. 1,288 Gemini 3.6 Flash
  18. 1,288 Ling-3.1-flash
  19. 1,286 DeepSeek V4.1 Flash
  20. 1,283 GLM-5.2
  21. 1,282 Qwen 3.8 Max
  22. 1,282 Grok 4.6
  23. 1,278 Claude Haiku 5.5
  24. 1,277 Deca 2.5 Ultra
  25. 1,277 Claude Opus 4.6
  26. 1,277 Gemini 3.8 Flash
  27. 1,276 Claude Sonnet 4.6
  28. 1,276 Qwen3.7-Plus
  29. 1,275 GPT-6 Sol
  30. 1,274 Qwen3.7-Max
  31. 1,274 Mistral Large 4
  32. 1,273 GPT-6.1 Sol
  33. 1,265 Grok 4.5
  34. 1,264 Muse Spark 1.1
  35. 1,261 GLM-5-Turbo
  36. 1,260 GPT-5.5
  37. 1,258 GLM-5.1
  38. 1,258 Muse Spark
  39. 1,258 MiMo-V2.5-Pro
  40. 1,257 GLM-5.3-Flash
  41. 1,253 GLM-5.3
  42. 1,250 MiMo-V2.5
  43. 1,247 Kimi K2.6
  44. 1,247 Claude Sonnet 5
  45. 1,241 Claude Opus 4.5
  46. 1,238 Gemini 3.5 Flash
  47. 1,237 GPT-5.6 Terra
  48. 1,236 Gemini 3 Pro
  49. 1,235 Claude Opus 4.8
  50. 1,233 GPT-5.4
  51. 1,231 MiniMax-M3
  52. 1,230 MiniMax-M2.7
  53. 1,226 Deca 2.5 Mini
  54. 1,226 GLM-5
  55. 1,224 Qwen 3.6 Plus
  56. 1,224 Grok 4.7
  57. 1,223 Nex N2 Pro
  58. 1,222 Kimi K2.7 Code
  59. 1,220 Gemini 2.5 Pro
  60. 1,219 GPT-5
  61. 1,217 Kimi K2.5
  62. 1,217 MiniMax-M3.1-Flash
  63. 1,215 Grok 4.20 Beta
  64. 1,211 GPT-5.1-Codex
  65. 1,209 GPT-6 Luna
  66. 1,207 Gemini 3.1 Pro
  67. 1,207 Mercury 3
  68. 1,204 MiniMax-M2.1
  69. 1,204 GPT-5.6 Luna
  70. 1,202 GPT-5.1
  71. 1,201 DeepSeek-V4-Pro
  72. 1,200 GLM-5V-Turbo
  73. 1,194 Claude 3.7 Sonnet
  74. 1,193 GLM-4.7
  75. 1,184 GLM-4.5-Air
  76. 1,184 GPT-5.2
  77. 1,180 DeepSeek-R1
  78. 1,177 Grok 4.3
  79. 1,175 Solar Pro 4
  80. 1,173 Qwen3.5 397B-A17B
  81. 1,173 Step 3.7 Flash
  82. 1,170 DeepSeek V4 Flash 0731
  83. 1,169 MiniMax-M2.5
  84. 1,166 DeepSeek-V3.1
  85. 1,166 GPT-6.1 Terra
  86. 1,165 Claude Sonnet 4.5
  87. 1,164 Claude Opus 4.1
  88. 1,163 Tencent Hy3
  89. 1,162 GPT-5.3 Codex
  90. 1,159 GLM-4.6
  91. 1,158 Inkling
  92. 1,157 GLM-4.5
  93. 1,156 DeepSeek-V3.2
  94. 1,155 Claude Sonnet 4
  95. 1,150 DeepSeek-V3.2-Exp
  96. 1,144 Claude Opus 4
  97. 1,142 Mistral Medium 3.1
  98. 1,138 MiniMax M2 Stable
  99. 1,138 Nemotron 3 Ultra
  100. 1,135 MiMo-V2-Flash
  101. 1,134 Mistral Large 3
  102. 1,133 Qwen3.5 Plus 02-15
  103. 1,132 Gemini 2.5 Flash
  104. 1,128 GPT-5 mini (Default)
  105. 1,125 DeepSeek-V4-Flash
  106. 1,122 Claude Haiku 4.5
  107. 1,121 glm-4.7-flash
  108. 1,118 Mistral Medium 3.5
  109. 1,108 AesCoder-4B
  110. 1,098 Qwen3-Max
  111. 1,096 Trinity Large Thinking
  112. 1,095 Grok 4 Fast
  113. 1,095 GPT-5.1-Codex-mini
  114. 1,087 DeepSeek-V3
  115. 1,086 Grok 4.1 Fast
  116. 1,085 Nemotron 3.5 Super VL
  117. 1,082 AGI-01 Swift
  118. 1,076 Prime Intellect: INTELLECT-3
  119. 1,068 Qwen3-235B-A22B-Instruct-2507
  120. 1,054 GPT-5 nano (Default)
  121. 1,043 Gemini 3.1 Flash-Lite
  122. 1,032 Mistral Medium 3 (2505)
  123. 1,029 Grok 4
  124. 1,028 Grok 3
  125. 1,015 Codestral 2508
  126. 1,012 Mercury 2
  127. 989 o4-mini
  128. 985 gpt-oss-120b
  129. 983 v0-1.5-md
  130. 933 Devstral Medium
  131. 901 Olmo 3.1 32B Think
  132. 889 GPT-4.1 nano
  133. 875 Grok-3 mini

Bars start at 800.

  1. 1 Claude Opus 5.5 1,388
  2. 2 Claude Sonnet 5.5 1,360
  3. 3 Kimi K3 1,351
  4. 4 Claude Fable 5.1 1,346
  5. 5 Muse Spark 1.3 xhigh 1,343
  6. 6 GPT-6 Astra xhigh, max 1,337
  7. 7 MiMo-V2.6 Pro 1,337
  8. 8 Muse Spark 1.2 1,330
  9. 9 Claude Opus 5 1,327
  10. 10 Gemini 4 Argon 1,325
  11. 11 GPT-5.6 Sol medium, xhigh 1,322
  12. 12 Muse Ember 1,317
  13. 13 Gemini 3.7 Flash 1,310
  14. 14 Claude Fable 5 1,304
  15. 15 Kimi K3.1 Pro 1,304
  16. 16 Muse Wonder max 1,290
  17. 17 Gemini 3.6 Flash 1,288
  18. 18 Ling-3.1-flash 1,288
  19. 19 DeepSeek V4.1 Flash 1,286
  20. 20 GLM-5.2 1,283
  21. 21 Qwen 3.8 Max 1,282
  22. 22 Grok 4.6 1,282
  23. 23 Claude Haiku 5.5 1,278
  24. 24 Deca 2.5 Ultra 2 1,277
  25. 25 Claude Opus 4.6 1,277
  26. 26 Gemini 3.8 Flash 1,277
  27. 27 Claude Sonnet 4.6 1,276
  28. 28 Qwen3.7-Plus 1,276
  29. 29 GPT-6 Sol 1,275
  30. 30 Qwen3.7-Max 1,274
  31. 31 Mistral Large 4 1,274
  32. 32 GPT-6.1 Sol 1,273
  33. 33 Grok 4.5 1,265
  34. 34 Muse Spark 1.1 1,264
  35. 35 GLM-5-Turbo 1,261
  36. 36 GPT-5.5 1,260
  37. 37 GLM-5.1 1,258
  38. 38 Muse Spark 1,258
  39. 39 MiMo-V2.5-Pro 1,258
  40. 40 GLM-5.3-Flash 1,257
  41. 41 GLM-5.3 1,253
  42. 42 MiMo-V2.5 1,250
  43. 43 Kimi K2.6 1,247
  44. 44 Claude Sonnet 5 1,247
  45. 45 Claude Opus 4.5 1,241
  46. 46 Gemini 3.5 Flash 1,238
  47. 47 GPT-5.6 Terra 1,237
  48. 48 Gemini 3 Pro 1,236
  49. 49 Claude Opus 4.8 1,235
  50. 50 GPT-5.4 medium 1,233
  51. 51 MiniMax-M3 1,231
  52. 52 MiniMax-M2.7 1,230
  53. 53 Deca 2.5 Mini 1,226
  54. 54 GLM-5 1,226
  55. 55 Qwen 3.6 Plus 1,224
  56. 56 Grok 4.7 1,224
  57. 57 Nex N2 Pro 1,223
  58. 58 Kimi K2.7 Code 1,222
  59. 59 Gemini 2.5 Pro Jun 2025 1,220
  60. 60 GPT-5 minimal, high 1,219
  61. 61 Kimi K2.5 thinking 1,217
  62. 62 MiniMax-M3.1-Flash 1,217
  63. 63 Grok 4.20 Beta reasoning 1,215
  64. 64 GPT-5.1-Codex 1,211
  65. 65 GPT-6 Luna 1,209
  66. 66 Gemini 3.1 Pro 1,207
  67. 67 Mercury 3 1,207
  68. 68 MiniMax-M2.1 1,204
  69. 69 GPT-5.6 Luna 1,204
  70. 70 GPT-5.1 low 1,202
  71. 71 DeepSeek-V4-Pro 1,201
  72. 72 GLM-5V-Turbo 1,200
  73. 73 Claude 3.7 Sonnet 1,194
  74. 74 GLM-4.7 1,193
  75. 75 GLM-4.5-Air 1,184
  76. 76 GPT-5.2 low 1,184
  77. 77 DeepSeek-R1 May 2025 1,180
  78. 78 Grok 4.3 1,177
  79. 79 Solar Pro 4 1,175
  80. 80 Qwen3.5 397B-A17B 1,173
  81. 81 Step 3.7 Flash 1,173
  82. 82 DeepSeek V4 Flash 0731 1,170
  83. 83 MiniMax-M2.5 1,169
  84. 84 DeepSeek-V3.1 thinking 1,166
  85. 85 GPT-6.1 Terra 1,166
  86. 86 Claude Sonnet 4.5 1,165
  87. 87 Claude Opus 4.1 thinking 1,164
  88. 88 Tencent Hy3 1,163
  89. 89 GPT-5.3 Codex 1,162
  90. 90 GLM-4.6 1,159
  91. 91 Inkling 1,158
  92. 92 GLM-4.5 1,157
  93. 93 DeepSeek-V3.2 1,156
  94. 94 Claude Sonnet 4 thinking 1,155
  95. 95 DeepSeek-V3.2-Exp 1,150
  96. 96 Claude Opus 4 1,144
  97. 97 Mistral Medium 3.1 1,142
  98. 98 MiniMax M2 Stable 1,138
  99. 99 Nemotron 3 Ultra 1,138
  100. 100 MiMo-V2-Flash 1,135
  101. 101 Mistral Large 3 1,134
  102. 102 Qwen3.5 Plus 02-15 1,133
  103. 103 Gemini 2.5 Flash Sep 2025 1,132
  104. 104 GPT-5 mini (Default) 1,128
  105. 105 DeepSeek-V4-Flash 1,125
  106. 106 Claude Haiku 4.5 1,122
  107. 107 glm-4.7-flash 1,121
  108. 108 Mistral Medium 3.5 1,118
  109. 109 AesCoder-4B 1,108
  110. 110 Qwen3-Max 1,098
  111. 111 Trinity Large Thinking 1,096
  112. 112 Grok 4 Fast 1,095
  113. 113 GPT-5.1-Codex-mini 1,095
  114. 114 DeepSeek-V3 Mar 2025 1,087
  115. 115 Grok 4.1 Fast reasoning 1,086
  116. 116 Nemotron 3.5 Super VL 1,085
  117. 117 AGI-01 Swift 1,082
  118. 118 Prime Intellect: INTELLECT-3 1,076
  119. 119 Qwen3-235B-A22B-Instruct-2507 1,068
  120. 120 GPT-5 nano (Default) 1,054
  121. 121 Gemini 3.1 Flash-Lite 1,043
  122. 122 Mistral Medium 3 (2505) 1,032
  123. 123 Grok 4 1,029
  124. 124 Grok 3 1,028
  125. 125 Codestral 2508 1,015
  126. 126 Mercury 2 1,012
  127. 127 o4-mini 989
  128. 128 gpt-oss-120b 985
  129. 129 v0-1.5-md 983
  130. 130 Devstral Medium 933
  131. 131 Olmo 3.1 32B Think 901
  132. 132 GPT-4.1 nano 889
  133. 133 Grok-3 mini 875

Source: Official DesignArena Data Visualization leaderboard

#EntryScoreSettingDateSource
1 Claude Opus 5.5 1,388 Sep 2026 Official leaderboard
2 Claude Sonnet 5.5 1,360 Sep 2026 Official leaderboard
3 Kimi K3 1,351 Jul 2026 Official leaderboard
4 Claude Fable 5.1 1,346 Sep 2026 Official leaderboard
5 Muse Spark 1.3 1,343 xhigh Sep 2026 Official leaderboard
6 GPT-6 Astra 1,337 xhigh, max Sep 2026 Official leaderboard
7 MiMo-V2.6 Pro 1,337 Sep 2026 Official leaderboard
8 Muse Spark 1.2 1,330 Aug 2026 Official leaderboard
9 Claude Opus 5 1,327 Jul 2026 Official leaderboard
10 Gemini 4 Argon 1,325 Sep 2026 Official leaderboard
11 GPT-5.6 Sol 1,322 medium, xhigh Jul 2026 Official leaderboard
12 Muse Ember 1,317 Nov 2026 Official leaderboard
13 Gemini 3.7 Flash 1,310 Aug 2026 Official leaderboard
14 Claude Fable 5 1,304 Jun 2026 Official leaderboard
15 Kimi K3.1 Pro 1,304 Oct 2026 Official leaderboard
16 Muse Wonder 1,290 max Oct 2026 Official leaderboard
17 Gemini 3.6 Flash 1,288 Jul 2026 Official leaderboard
18 Ling-3.1-flash 1,288 Oct 2026 Official leaderboard
19 DeepSeek V4.1 Flash 1,286 Sep 2026 Official leaderboard
20 GLM-5.2 1,283 Jun 2026 Official leaderboard
21 Qwen 3.8 Max 1,282 Aug 2026 Official leaderboard
22 Grok 4.6 1,282 Aug 2026 Official leaderboard
23 Claude Haiku 5.5 1,278 Oct 2026 Official leaderboard
24 Deca 2.5 Ultra 1,277 2 Official leaderboard
25 Claude Opus 4.6 1,277 Feb 2026 Official leaderboard
26 Gemini 3.8 Flash 1,277 Sep 2026 Official leaderboard
27 Claude Sonnet 4.6 1,276 Feb 2026 Official leaderboard
28 Qwen3.7-Plus 1,276 Jun 2026 Official leaderboard
29 GPT-6 Sol 1,275 Sep 2026 Official leaderboard
30 Qwen3.7-Max 1,274 May 2026 Official leaderboard
31 Mistral Large 4 1,274 Oct 2026 Official leaderboard
32 GPT-6.1 Sol 1,273 Sep 2026 Official leaderboard
33 Grok 4.5 1,265 Jul 2026 Official leaderboard
34 Muse Spark 1.1 1,264 Jul 2026 Official leaderboard
35 GLM-5-Turbo 1,261 Mar 2026 Official leaderboard
36 GPT-5.5 1,260 Apr 2026 Official leaderboard
37 GLM-5.1 1,258 Apr 2026 Official leaderboard
38 Muse Spark 1,258 Apr 2026 Official leaderboard
39 MiMo-V2.5-Pro 1,258 Apr 2026 Official leaderboard
40 GLM-5.3-Flash 1,257 Aug 2026 Official leaderboard
41 GLM-5.3 1,253 Aug 2026 Official leaderboard
42 MiMo-V2.5 1,250 Apr 2026 Official leaderboard
43 Kimi K2.6 1,247 Apr 2026 Official leaderboard
44 Claude Sonnet 5 1,247 Jun 2026 Official leaderboard
45 Claude Opus 4.5 1,241 Nov 2025 Official leaderboard
46 Gemini 3.5 Flash 1,238 May 2026 Official leaderboard
47 GPT-5.6 Terra 1,237 Jul 2026 Official leaderboard
48 Gemini 3 Pro 1,236 Nov 2025 Official leaderboard
49 Claude Opus 4.8 1,235 May 2026 Official leaderboard
50 GPT-5.4 1,233 medium Mar 2026 Official leaderboard
51 MiniMax-M3 1,231 Jun 2026 Official leaderboard
52 MiniMax-M2.7 1,230 Mar 2026 Official leaderboard
53 Deca 2.5 Mini 1,226 Official leaderboard
54 GLM-5 1,226 Feb 2026 Official leaderboard
55 Qwen 3.6 Plus 1,224 Mar 2026 Official leaderboard
56 Grok 4.7 1,224 Sep 2026 Official leaderboard
57 Nex N2 Pro 1,223 Official leaderboard
58 Kimi K2.7 Code 1,222 Jun 2026 Official leaderboard
59 Gemini 2.5 Pro Jun 2025 1,220 Jun 2025 Official leaderboard
60 GPT-5 1,219 minimal, high Aug 2025 Official leaderboard
61 Kimi K2.5 1,217 thinking Jan 2026 Official leaderboard
62 MiniMax-M3.1-Flash 1,217 Oct 2026 Official leaderboard
63 Grok 4.20 Beta 1,215 reasoning Official leaderboard
64 GPT-5.1-Codex 1,211 Nov 2025 Official leaderboard
65 GPT-6 Luna 1,209 Sep 2026 Official leaderboard
66 Gemini 3.1 Pro 1,207 Feb 2026 Official leaderboard
67 Mercury 3 1,207 Oct 2026 Official leaderboard
68 MiniMax-M2.1 1,204 Dec 2025 Official leaderboard
69 GPT-5.6 Luna 1,204 Jul 2026 Official leaderboard
70 GPT-5.1 1,202 low Nov 2025 Official leaderboard
71 DeepSeek-V4-Pro 1,201 Apr 2026 Official leaderboard
72 GLM-5V-Turbo 1,200 Apr 2026 Official leaderboard
73 Claude 3.7 Sonnet 1,194 Feb 2025 Official leaderboard
74 GLM-4.7 1,193 Dec 2025 Official leaderboard
75 GLM-4.5-Air 1,184 Jul 2025 Official leaderboard
76 GPT-5.2 1,184 low Dec 2025 Official leaderboard
77 DeepSeek-R1 May 2025 1,180 May 2025 Official leaderboard
78 Grok 4.3 1,177 Apr 2026 Official leaderboard
79 Solar Pro 4 1,175 Aug 2026 Official leaderboard
80 Qwen3.5 397B-A17B 1,173 Feb 2026 Official leaderboard
81 Step 3.7 Flash 1,173 May 2026 Official leaderboard
82 DeepSeek V4 Flash 0731 1,170 Jul 2026 Official leaderboard
83 MiniMax-M2.5 1,169 Feb 2026 Official leaderboard
84 DeepSeek-V3.1 1,166 thinking Aug 2025 Official leaderboard
85 GPT-6.1 Terra 1,166 Oct 2026 Official leaderboard
86 Claude Sonnet 4.5 1,165 Sep 2025 Official leaderboard
87 Claude Opus 4.1 1,164 thinking Aug 2025 Official leaderboard
88 Tencent Hy3 1,163 Jul 2026 Official leaderboard
89 GPT-5.3 Codex 1,162 Feb 2026 Official leaderboard
90 GLM-4.6 1,159 Sep 2025 Official leaderboard
91 Inkling 1,158 Jul 2026 Official leaderboard
92 GLM-4.5 1,157 Aug 2025 Official leaderboard
93 DeepSeek-V3.2 1,156 Dec 2025 Official leaderboard
94 Claude Sonnet 4 1,155 thinking May 2025 Official leaderboard
95 DeepSeek-V3.2-Exp 1,150 Sep 2025 Official leaderboard
96 Claude Opus 4 1,144 May 2025 Official leaderboard
97 Mistral Medium 3.1 1,142 Aug 2025 Official leaderboard
98 MiniMax M2 Stable 1,138 Official leaderboard
99 Nemotron 3 Ultra 1,138 Jun 2026 Official leaderboard
100 MiMo-V2-Flash 1,135 Dec 2025 Official leaderboard
101 Mistral Large 3 1,134 Dec 2025 Official leaderboard
102 Qwen3.5 Plus 02-15 1,133 Official leaderboard
103 Gemini 2.5 Flash Sep 2025 1,132 Sep 2025 Official leaderboard
104 GPT-5 mini (Default) 1,128 Official leaderboard
105 DeepSeek-V4-Flash 1,125 Apr 2026 Official leaderboard
106 Claude Haiku 4.5 1,122 Oct 2025 Official leaderboard
107 glm-4.7-flash 1,121 Jan 2026 Official leaderboard
108 Mistral Medium 3.5 1,118 Apr 2026 Official leaderboard
109 AesCoder-4B 1,108 Official leaderboard
110 Qwen3-Max 1,098 Sep 2025 Official leaderboard
111 Trinity Large Thinking 1,096 Apr 2026 Official leaderboard
112 Grok 4 Fast 1,095 Sep 2025 Official leaderboard
113 GPT-5.1-Codex-mini 1,095 Nov 2025 Official leaderboard
114 DeepSeek-V3 Mar 2025 1,087 Mar 2025 Official leaderboard
115 Grok 4.1 Fast 1,086 reasoning Nov 2025 Official leaderboard
116 Nemotron 3.5 Super VL 1,085 Oct 2026 Official leaderboard
117 AGI-01 Swift 1,082 Official leaderboard
118 Prime Intellect: INTELLECT-3 1,076 Official leaderboard
119 Qwen3-235B-A22B-Instruct-2507 1,068 Official leaderboard
120 GPT-5 nano (Default) 1,054 Official leaderboard
121 Gemini 3.1 Flash-Lite 1,043 Mar 2026 Official leaderboard
122 Mistral Medium 3 (2505) 1,032 Official leaderboard
123 Grok 4 1,029 Jul 2025 Official leaderboard
124 Grok 3 1,028 Apr 2025 Official leaderboard
125 Codestral 2508 1,015 Official leaderboard
126 Mercury 2 1,012 Feb 2026 Official leaderboard
127 o4-mini 989 Apr 2025 Official leaderboard
128 gpt-oss-120b 985 Aug 2025 Official leaderboard
129 v0-1.5-md 983 Official leaderboard
130 Devstral Medium 933 Jul 2025 Official leaderboard
131 Olmo 3.1 32B Think 901 Dec 2025 Official leaderboard
132 GPT-4.1 nano 889 Apr 2025 Official leaderboard
133 Grok-3 mini 875 Apr 2025 Official leaderboard

More effort adds a median 31 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
Muse Spark 1.3
xhigh, max xhigh
1,343
GPT-6 Astra
xhigh
1,337
GPT-5.6 Sol
medium medium, xhi…
1,322
DeepSeek V4.1 Flash
1017
1,286
Claude Opus 4.6
thinking
1,277
GPT-5.4
none medium
1,233
GPT-5
minimal minimal, hi…
1,219
GPT-5.1
none
1,202
GPT-5.2
none medium
1,184
DeepSeek-V3.1
thinking
1,166
Claude Sonnet 4.5
thinking
1,165
Claude Opus 4.1
thinking
1,164

Source: Official DesignArena Data Visualization leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 1,388
Claude Sonnet 5.5 1,360
Kimi K3 1,351
Claude Fable 5.1 1,346
Muse Spark 1.3 xhigh 1,343
GPT-6 Astra xhigh, max 1,337
MiMo-V2.6 Pro 1,337
Muse Spark 1.2 1,330
Claude Opus 5 1,327
Gemini 4 Argon 1,325
GPT-5.6 Sol medium, xhigh 1,322
GPT-6 Astra xhigh 1,318
Muse Ember 1,317
Gemini 3.7 Flash 1,310
Muse Spark 1.3 xhigh, max 1,308
Claude Fable 5 1,304
Kimi K3.1 Pro 1,304
GPT-5.6 Sol medium 1,291
Muse Wonder max 1,290
Gemini 3.6 Flash 1,288
Ling-3.1-flash 1,288
DeepSeek V4.1 Flash 1,286
GLM-5.2 1,283
Qwen 3.8 Max 1,282
Grok 4.6 1,282
Claude Haiku 5.5 1,278
Claude Opus 4.6 1,277
Gemini 3.8 Flash 1,277
Claude Sonnet 4.6 1,276
Qwen3.7-Plus 1,276
GPT-6 Sol 1,275
Qwen3.7-Max 1,274
Mistral Large 4 1,274
GPT-6.1 Sol 1,273
Claude Opus 4.6 thinking 1,268
Grok 4.5 1,265
Muse Spark 1.1 1,264
GLM-5-Turbo 1,261
GPT-5.5 1,260
DeepSeek V4.1 Flash 1017 1,259
GLM-5.1 1,258
Muse Spark 1,258
MiMo-V2.5-Pro 1,258
GLM-5.3-Flash 1,257
GLM-5.3 1,253
MiMo-V2.5 1,250
Kimi K2.6 1,247
Claude Sonnet 5 1,247
Claude Opus 4.5 1,241
Gemini 3.5 Flash 1,238
GPT-5.6 Terra 1,237
Gemini 3 Pro 1,236
Claude Opus 4.8 1,235
GPT-5.4 medium 1,233
MiniMax-M3 1,231
MiniMax-M2.7 1,230
GLM-5 1,226
Qwen 3.6 Plus 1,224
Grok 4.7 1,224
Kimi K2.7 Code 1,222
Gemini 2.5 Pro Jun 2025 1,220
GPT-5 minimal, high 1,219
Kimi K2.5 thinking 1,217
MiniMax-M3.1-Flash 1,217
GPT-5.1-Codex 1,211
GPT-6 Luna 1,209
Gemini 3.1 Pro 1,207
Mercury 3 1,207
MiniMax-M2.1 1,204
GPT-5.6 Luna 1,204
GPT-5.1 low 1,202
DeepSeek-V4-Pro 1,201
GLM-5V-Turbo 1,200
Claude 3.7 Sonnet 1,194
GPT-5.4 low 1,194
GPT-5.1 none 1,193
GLM-4.7 1,193
GPT-5.1 high 1,192
GPT-5.1 medium 1,190
GLM-4.5-Air 1,184
GPT-5.2 low 1,184
GPT-5.2 none 1,182
DeepSeek-R1 May 2025 1,180
GPT-5.4 none 1,179
Grok 4.3 1,177
Solar Pro 4 1,175
GPT-5 minimal 1,174
Qwen3.5 397B-A17B 1,173
Step 3.7 Flash 1,173
DeepSeek V4 Flash 0731 1,170
MiniMax-M2.5 1,169
DeepSeek-V3.1 thinking 1,166
GPT-6.1 Terra 1,166
Claude Sonnet 4.5 1,165
Claude Opus 4.1 thinking 1,164
Tencent Hy3 1,163
GPT-5.3 Codex 1,162
GLM-4.6 1,159
Claude Opus 4.1 1,158
Inkling 1,158
GLM-4.5 1,157
Claude Sonnet 4.5 thinking 1,156
DeepSeek-V3.2 1,156
Claude Sonnet 4 thinking 1,155
GPT-5.2 medium 1,154
DeepSeek-V3.2-Exp 1,150
Claude Opus 4 1,144
Mistral Medium 3.1 1,142
Nemotron 3 Ultra 1,138
MiMo-V2-Flash 1,135
Mistral Large 3 1,134
Gemini 2.5 Flash Sep 2025 1,132
DeepSeek-V4-Flash 1,125
Tencent Hy3 low 1,123
Claude Haiku 4.5 1,122
glm-4.7-flash 1,121
Mistral Medium 3.5 1,118
Claude Sonnet 4 1,110
Qwen3-Max 1,098
Trinity Large Thinking 1,096
Grok 4 Fast 1,095
GPT-5.1-Codex-mini 1,095
DeepSeek-V3.1 1,094
DeepSeek-V3 Mar 2025 1,087
Grok 4.1 Fast reasoning 1,086
Nemotron 3.5 Super VL 1,085
Grok 4.1 Fast 1,062
Gemini 2.5 Flash Sep 2025 1,058
Gemini 3.1 Flash-Lite 1,043
Grok 4 Fast reasoning 1,037
Grok 4 1,029
Grok 3 1,028
Mercury 2 1,012
o4-mini 989
gpt-oss-120b 985
Devstral Medium 933
Olmo 3.1 32B Think 901
GPT-4.1 nano 889
Grok-3 mini 875

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.

Anthropic Google DeepMind OpenAI Moonshot AI
Own leaderboard 9 leaders
Claude Opus 5.5 1,388

Source: Official DesignArena Data Visualization 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
800 1,000 1,200 1,400 2025 2026 2027 800 1,000 1,200 1,400 2025 2026 2027

Source: Official DesignArena Data Visualization 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