Benchmark Almanac

Code Arena

Claude Fable 5.2 1,836 leads since November 2026
Claude Fable 5.2 1,836 leads since November 2026
In use

Models act as coding agents that build web apps from user prompts, and users vote between two results. Votes yield Elo-style ratings.

It started a fresh leaderboard and does not reuse WebDev Arena votes. It splits results into Fullstack and Frontend categories, with HTML, React and domain filters. On October 27th, 2026 it listed 153 models and 952,039 votes. Image-to-WebDev is a separate board.

Released
November 2025
Kind
Web and app front ends
Scored
Ranked by votes
Answers
Open-ended
Strongest model tested
Claude Fable 5.2 #1 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
Scores refreshed
November 12th, 2026
133
Entries on its own leaderboard, 128 models we track
2
Launch posts list it, from 2 labs
3
System cards report it, from 3 labs

Claude Fable 5.2 leads with 1,836

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.

Claude Fable 5.2 leads with 1,836, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

133 entries. Scroll sideways for the rest. Bars start at 1,000.

  1. 1,836 Claude Fable 5.2
  2. 1,814 Claude Opus 5.5
  3. 1,773 Claude Sonnet 5.5
  4. 1,768 Gemini 4 Carbon
  5. 1,764 GPT-6 Astra
  6. 1,743 GPT-6.1 Sol
  7. 1,742 Claude Fable 5.1
  8. 1,723 Muse Wonder Full
  9. 1,691 Claude Opus 5
  10. 1,688 Kimi K3.1 Pro
  11. 1,682 Gemini 4 Argon
  12. 1,679 GPT-6.1 Terra
  13. 1,676 Qwen3.8 Max (0902)
  14. 1,673 GPT-6 Sol
  15. 1,670 Qwen 3.8 Max
  16. 1,664 DeepSeek V4.1 Flash
  17. 1,658 Kimi K3
  18. 1,655 Muse Spark 1.3
  19. 1,649 Muse Wonder
  20. 1,646 Inkling 2 Small
  21. 1,641 Grok 4.7
  22. 1,638 MiMo-V2.6 Flash
  23. 1,632 Tencent Hy4 preview
  24. 1,625 Claude Fable 5
  25. 1,624 MiMo-V2.6 Pro
  26. 1,624 Qwen3.8-Flash-Next
  27. 1,622 GLM-5.3
  28. 1,618 Grok 4.6
  29. 1,615 GPT-5.6 Sol
  30. 1,611 U2 Flash
  31. 1,609 GLM-5.3-Flash
  32. 1,609 Muse Ember
  33. 1,608 GLM-5.2
  34. 1,594 Claude Haiku 5.5
  35. 1,590 Qwen 3.8 27B
  36. 1,588 Gemini 3.7 Flash
  37. 1,585 MiniMax-M3.1-Flash
  38. 1,583 Gemini 3.8 Flash
  39. 1,582 DeepSeek-V4-Flash
  40. 1,581 DeepSeek-V4-Pro
  41. 1,577 GPT-6 Luna
  42. 1,573 Step 5 Preview
  43. 1,564 Mistral Large 4
  44. 1,557 Claude Opus 4.7
  45. 1,555 Claude Opus 4.8
  46. 1,552 Grok 4.5
  47. 1,547 Claude Opus 4.6
  48. 1,541 Muse Spark 1.1
  49. 1,539 Claude Sonnet 5
  50. 1,538 Gemini 3.6 Flash
  51. 1,534 Muse Spark 1.2
  52. 1,526 Muse Spark Open
  53. 1,522 Claude Sonnet 4.6
  54. 1,520 GPT-5.6 Terra
  55. 1,519 GPT-5.6 Luna
  56. 1,519 Seed2.1 Pro
  57. 1,515 Qwen3.7-Max
  58. 1,512 GPT-5.5
  59. 1,509 Kimi K2.6
  60. 1,508 GLM-5.1
  61. 1,506 Tencent Hy3
  62. 1,500 Gemini 3.5 Flash
  63. 1,493 Claude Opus 4.5
  64. 1,485 Spark-X2.6
  65. 1,484 MiniMax-M3
  66. 1,482 Qwen 3.6 Max (Preview)
  67. 1,479 MiMo-V2.5-Pro
  68. 1,474 Kimi K2.7 Code
  69. 1,462 GPT-5.4
  70. 1,461 Qwen 3.6 Plus
  71. 1,446 Gemini 3.1 Pro
  72. 1,441 Gemini 3 Pro
  73. 1,440 Gemini 3 Flash
  74. 1,438 MiMo-V2.5
  75. 1,437 Beam 1.0
  76. 1,436 Kimi K2.5
  77. 1,435 Gemini 3.5 Flash-Lite
  78. 1,435 GLM-5
  79. 1,435 GLM-4.7
  80. 1,434 MiMo-V2-Pro
  81. 1,420 GPT-5.2
  82. 1,417 GPT-5
  83. 1,414 Inkling
  84. 1,408 Inkling-Small
  85. 1,406 GPT-5.3 Codex
  86. 1,401 Nemotron 3.5 Super VL
  87. 1,400 Qwen3.5 397B-A17B
  88. 1,399 GLM-5V-Turbo
  89. 1,398 MiniMax-M2.7
  90. 1,398 GPT-5.4 Mini
  91. 1,397 GPT-5.1
  92. 1,393 Claude Sonnet 4.5
  93. 1,389 Claude Opus 4.1
  94. 1,386 MiniMax-M2.5
  95. 1,384 MiniMax-M2.1
  96. 1,375 Grok 4.20 Beta (Reasoning, 0309)
  97. 1,370 Gemma 4 31B IT
  98. 1,369 Solar Pro 4
  99. 1,362 DeepSeek-V3.2
  100. 1,360 Qwen3.5-122B-A10B
  101. 1,358 Gemma 4 26B A4B
  102. 1,358 Tencent Hy3 preview
  103. 1,357 Grok 4.3
  104. 1,357 Qwen3.5-27B
  105. 1,355 Muse Glimmer
  106. 1,350 Laguna M.1
  107. 1,342 GLM-4.6
  108. 1,338 GPT-5.2 Codex
  109. 1,338 GPT-5.1-Codex
  110. 1,331 MiMo-V2-Flash
  111. 1,330 Claude Haiku 4.5
  112. 1,322 Kimi K2 Thinking Turbo
  113. 1,303 Laguna XS.2
  114. 1,299 MiniMax-M2
  115. 1,275 Qwen3-Coder-480B-A35B
  116. 1,270 DeepSeek-V3.2-Exp
  117. 1,263 Mistral Medium 3.5
  118. 1,258 KAT-Coder-Pro V1
  119. 1,256 Gemini 3.1 Flash-Lite
  120. 1,253 Qwen3.5-35B-A3B
  121. 1,245 GPT-5.1-Codex-mini
  122. 1,244 Qwen3.5-Flash
  123. 1,242 Grok 4.1 Fast
  124. 1,236 Trinity Large Thinking
  125. 1,233 Mistral Large 3
  126. 1,228 Gemini 2.5 Pro
  127. 1,217 Grok 4.1
  128. 1,196 Devstral 2
  129. 1,193 Granite 4.1 8B
  130. 1,171 Mercury 2
  131. 1,164 grok-code-fast-1
  132. 1,162 Grok 4 Fast
  133. 1,072 Devstral Medium

Bars start at 1,000.

  1. 1 Claude Fable 5.2 xhigh 1,836
  2. 2 Claude Opus 5.5 max 1,814
  3. 3 Claude Sonnet 5.5 xhigh 1,773
  4. 4 Gemini 4 Carbon high 1,768
  5. 5 GPT-6 Astra max 1,764
  6. 6 GPT-6.1 Sol max 1,743
  7. 7 Claude Fable 5.1 max 1,742
  8. 8 Muse Wonder Full max 1,723
  9. 9 Claude Opus 5 max 1,691
  10. 10 Kimi K3.1 Pro max 1,688
  11. 11 Gemini 4 Argon high 1,682
  12. 12 GPT-6.1 Terra max 1,679
  13. 13 Qwen3.8 Max (0902) 1,676
  14. 14 GPT-6 Sol max 1,673
  15. 15 Qwen 3.8 Max 1,670
  16. 16 DeepSeek V4.1 Flash max 1,664
  17. 17 Kimi K3 max 1,658
  18. 18 Muse Spark 1.3 max 1,655
  19. 19 Muse Wonder max 1,649
  20. 20 Inkling 2 Small 1,646
  21. 21 Grok 4.7 xhigh 1,641
  22. 22 MiMo-V2.6 Flash 1,638
  23. 23 Tencent Hy4 preview 1,632
  24. 24 Claude Fable 5 high 1,625
  25. 25 MiMo-V2.6 Pro 1,624
  26. 26 Qwen3.8-Flash-Next 1,624
  27. 27 GLM-5.3 max 1,622
  28. 28 Grok 4.6 high 1,618
  29. 29 GPT-5.6 Sol 1,615
  30. 30 U2 Flash 1,611
  31. 31 GLM-5.3-Flash 1,609
  32. 32 Muse Ember max 1,609
  33. 33 GLM-5.2 max 1,608
  34. 34 Claude Haiku 5.5 high 1,594
  35. 35 Qwen 3.8 27B 1,590
  36. 36 Gemini 3.7 Flash high 1,588
  37. 37 MiniMax-M3.1-Flash 1,585
  38. 38 Gemini 3.8 Flash high 1,583
  39. 39 DeepSeek-V4-Flash high 1,582
  40. 40 DeepSeek-V4-Pro high 1,581
  41. 41 GPT-6 Luna max 1,577
  42. 42 Step 5 Preview high 1,573
  43. 43 Mistral Large 4 1,564
  44. 44 Claude Opus 4.7 high 1,557
  45. 45 Claude Opus 4.8 high 1,555
  46. 46 Grok 4.5 1,552
  47. 47 Claude Opus 4.6 high 1,547
  48. 48 Muse Spark 1.1 1,541
  49. 49 Claude Sonnet 5 high 1,539
  50. 50 Gemini 3.6 Flash high 1,538
  51. 51 Muse Spark 1.2 xhigh 1,534
  52. 52 Muse Spark Open max 1,526
  53. 53 Claude Sonnet 4.6 1,522
  54. 54 GPT-5.6 Terra 1,520
  55. 55 GPT-5.6 Luna 1,519
  56. 56 Seed2.1 Pro 1,519
  57. 57 Qwen3.7-Max 1,515
  58. 58 GPT-5.5 1,512
  59. 59 Kimi K2.6 1,509
  60. 60 GLM-5.1 1,508
  61. 61 Tencent Hy3 1,506
  62. 62 Gemini 3.5 Flash high 1,500
  63. 63 Claude Opus 4.5 high, 32k 1,493
  64. 64 Spark-X2.6 1,485
  65. 65 MiniMax-M3 1,484
  66. 66 Qwen 3.6 Max (Preview) 1,482
  67. 67 MiMo-V2.5-Pro 1,479
  68. 68 Kimi K2.7 Code 1,474
  69. 69 GPT-5.4 1,462
  70. 70 Qwen 3.6 Plus 1,461
  71. 71 Gemini 3.1 Pro 1,446
  72. 72 Gemini 3 Pro 1,441
  73. 73 Gemini 3 Flash 1,440
  74. 74 MiMo-V2.5 1,438
  75. 75 Beam 1.0 1,437
  76. 76 Kimi K2.5 thinking 1,436
  77. 77 Gemini 3.5 Flash-Lite 1,435
  78. 78 GLM-5 1,435
  79. 79 GLM-4.7 1,435
  80. 80 MiMo-V2-Pro 1,434
  81. 81 GPT-5.2 1,420
  82. 82 GPT-5 medium 1,417
  83. 83 Inkling 1,414
  84. 84 Inkling-Small 1,408
  85. 85 GPT-5.3 Codex 1,406
  86. 86 Nemotron 3.5 Super VL nvfp4 1,401
  87. 87 Qwen3.5 397B-A17B 1,400
  88. 88 GLM-5V-Turbo 1,399
  89. 89 MiniMax-M2.7 1,398
  90. 90 GPT-5.4 Mini high 1,398
  91. 91 GPT-5.1 medium 1,397
  92. 92 Claude Sonnet 4.5 high, 32k 1,393
  93. 93 Claude Opus 4.1 1,389
  94. 94 MiniMax-M2.5 1,386
  95. 95 MiniMax-M2.1 1,384
  96. 96 Grok 4.20 Beta (Reasoning, 0309) 1,375
  97. 97 Gemma 4 31B IT 1,370
  98. 98 Solar Pro 4 1,369
  99. 99 DeepSeek-V3.2 thinking 1,362
  100. 100 Qwen3.5-122B-A10B 1,360
  101. 101 Gemma 4 26B A4B 1,358
  102. 102 Tencent Hy3 preview 1,358
  103. 103 Grok 4.3 1,357
  104. 104 Qwen3.5-27B 1,357
  105. 105 Muse Glimmer 1,355
  106. 106 Laguna M.1 1,350
  107. 107 GLM-4.6 1,342
  108. 108 GPT-5.2 Codex 1,338
  109. 109 GPT-5.1-Codex 1,338
  110. 110 MiMo-V2-Flash non-thinking 1,331
  111. 111 Claude Haiku 4.5 1,330
  112. 112 Kimi K2 Thinking Turbo 1,322
  113. 113 Laguna XS.2 1,303
  114. 114 MiniMax-M2 1,299
  115. 115 Qwen3-Coder-480B-A35B 1,275
  116. 116 DeepSeek-V3.2-Exp 1,270
  117. 117 Mistral Medium 3.5 1,263
  118. 118 KAT-Coder-Pro V1 1,258
  119. 119 Gemini 3.1 Flash-Lite 1,256
  120. 120 Qwen3.5-35B-A3B 1,253
  121. 121 GPT-5.1-Codex-mini 1,245
  122. 122 Qwen3.5-Flash 1,244
  123. 123 Grok 4.1 Fast reasoning 1,242
  124. 124 Trinity Large Thinking 1,236
  125. 125 Mistral Large 3 1,233
  126. 126 Gemini 2.5 Pro Jun 2025 1,228
  127. 127 Grok 4.1 thinking 1,217
  128. 128 Devstral 2 1,196
  129. 129 Granite 4.1 8B 1,193
  130. 130 Mercury 2 1,171
  131. 131 grok-code-fast-1 1,164
  132. 132 Grok 4 Fast reasoning 1,162
  133. 133 Devstral Medium 1,072

Source: Official Code Arena leaderboard

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

  1. 1,836 Claude Fable 5.2
  2. 1,814 Claude Opus 5.5
  3. 1,773 Claude Sonnet 5.5
  4. 1,768 Gemini 4 Carbon
  5. 1,764 GPT-6 Astra
  6. 1,743 GPT-6.1 Sol
  7. 1,742 Claude Fable 5.1
  8. 1,723 Muse Wonder Full
  9. 1,691 Claude Opus 5
  10. 1,688 Kimi K3.1 Pro
  11. 1,682 Gemini 4 Argon
  12. 1,679 GPT-6.1 Terra
  13. 1,678 Kimi K3
  14. 1,676 Qwen3.8 Max (0902)
  15. 1,673 GPT-6 Sol
  16. 1,670 Qwen 3.8 Max
  17. 1,664 DeepSeek V4.1 Flash
  18. 1,655 Muse Spark 1.3
  19. 1,653 Inkling 2 Small
  20. 1,649 Muse Wonder
  21. 1,641 Grok 4.7
  22. 1,638 MiMo-V2.6 Flash
  23. 1,632 Tencent Hy4 preview
  24. 1,625 Claude Fable 5
  25. 1,624 MiMo-V2.6 Pro
  26. 1,624 Qwen3.8-Flash-Next
  27. 1,622 GLM-5.3
  28. 1,618 Grok 4.6
  29. 1,615 GPT-5.6 Sol
  30. 1,609 GLM-5.3-Flash
  31. 1,609 Muse Ember
  32. 1,608 GLM-5.2
  33. 1,594 Claude Haiku 5.5
  34. 1,590 Qwen 3.8 27B
  35. 1,588 Gemini 3.7 Flash
  36. 1,585 MiniMax-M3.1-Flash
  37. 1,583 Gemini 3.8 Flash
  38. 1,582 DeepSeek-V4-Flash
  39. 1,581 DeepSeek-V4-Pro
  40. 1,577 GPT-6 Luna
  41. 1,573 Step 5 Preview
  42. 1,564 Mistral Large 4
  43. 1,557 Claude Opus 4.7
  44. 1,555 Claude Opus 4.8
  45. 1,552 Grok 4.5
  46. 1,547 Claude Opus 4.6
  47. 1,541 Muse Spark 1.1
  48. 1,539 Claude Sonnet 5
  49. 1,538 Gemini 3.6 Flash
  50. 1,534 Muse Spark 1.2
  51. 1,526 Muse Spark Open
  52. 1,522 Claude Sonnet 4.6
  53. 1,520 GPT-5.6 Terra
  54. 1,519 GPT-5.6 Luna
  55. 1,519 Seed2.1 Pro
  56. 1,515 Qwen3.7-Max
  57. 1,512 GPT-5.5
  58. 1,509 Kimi K2.6
  59. 1,508 GLM-5.1
  60. 1,506 Tencent Hy3
  61. 1,500 Gemini 3.5 Flash
  62. 1,493 Claude Opus 4.5
  63. 1,484 MiniMax-M3
  64. 1,482 Qwen 3.6 Max (Preview)
  65. 1,479 MiMo-V2.5-Pro
  66. 1,474 Kimi K2.7 Code
  67. 1,462 GPT-5.4
  68. 1,461 Qwen 3.6 Plus
  69. 1,449 GLM-5
  70. 1,446 Gemini 3.1 Pro
  71. 1,441 Gemini 3 Pro
  72. 1,440 Gemini 3 Flash
  73. 1,438 MiMo-V2.5
  74. 1,437 Beam 1.0
  75. 1,436 Kimi K2.5
  76. 1,435 Gemini 3.5 Flash-Lite
  77. 1,435 GLM-4.7
  78. 1,434 MiMo-V2-Pro
  79. 1,420 GPT-5.2
  80. 1,417 GPT-5
  81. 1,414 Inkling
  82. 1,408 Inkling-Small
  83. 1,406 GPT-5.3 Codex
  84. 1,401 Nemotron 3.5 Super VL
  85. 1,400 Qwen3.5 397B-A17B
  86. 1,399 GLM-5V-Turbo
  87. 1,398 MiniMax-M2.7
  88. 1,398 GPT-5.4 Mini
  89. 1,397 GPT-5.1
  90. 1,393 Claude Sonnet 4.5
  91. 1,389 Claude Opus 4.1
  92. 1,386 MiniMax-M2.5
  93. 1,384 MiniMax-M2.1
  94. 1,370 Gemma 4 31B IT
  95. 1,369 Solar Pro 4
  96. 1,362 DeepSeek-V3.2
  97. 1,360 Qwen3.5-122B-A10B
  98. 1,358 Gemma 4 26B A4B
  99. 1,358 Tencent Hy3 preview
  100. 1,357 Grok 4.3
  101. 1,357 Qwen3.5-27B
  102. 1,355 Muse Glimmer
  103. 1,350 Laguna M.1
  104. 1,342 GLM-4.6
  105. 1,338 GPT-5.2 Codex
  106. 1,338 GPT-5.1-Codex
  107. 1,331 MiMo-V2-Flash
  108. 1,330 Claude Haiku 4.5
  109. 1,303 Laguna XS.2
  110. 1,299 MiniMax-M2
  111. 1,275 Qwen3-Coder-480B-A35B
  112. 1,270 DeepSeek-V3.2-Exp
  113. 1,263 Mistral Medium 3.5
  114. 1,256 Gemini 3.1 Flash-Lite
  115. 1,253 Qwen3.5-35B-A3B
  116. 1,245 GPT-5.1-Codex-mini
  117. 1,244 Qwen3.5-Flash
  118. 1,242 Grok 4.1 Fast
  119. 1,236 Trinity Large Thinking
  120. 1,233 Mistral Large 3
  121. 1,228 Gemini 2.5 Pro
  122. 1,217 Grok 4.1
  123. 1,196 Devstral 2
  124. 1,193 Granite 4.1 8B
  125. 1,171 Mercury 2
  126. 1,164 grok-code-fast-1
  127. 1,162 Grok 4 Fast
  128. 1,072 Devstral Medium

Bars start at 1,000.

  1. 1 Claude Fable 5.2 xhigh 1,836
  2. 2 Claude Opus 5.5 max 1,814
  3. 3 Claude Sonnet 5.5 xhigh 1,773
  4. 4 Gemini 4 Carbon high 1,768
  5. 5 GPT-6 Astra max 1,764
  6. 6 GPT-6.1 Sol max 1,743
  7. 7 Claude Fable 5.1 max 1,742
  8. 8 Muse Wonder Full max 1,723
  9. 9 Claude Opus 5 max 1,691
  10. 10 Kimi K3.1 Pro max 1,688
  11. 11 Gemini 4 Argon high 1,682
  12. 12 GPT-6.1 Terra max 1,679
  13. 13 Kimi K3 1,678
  14. 14 Qwen3.8 Max (0902) 1,676
  15. 15 GPT-6 Sol max 1,673
  16. 16 Qwen 3.8 Max 1,670
  17. 17 DeepSeek V4.1 Flash max 1,664
  18. 18 Muse Spark 1.3 max 1,655
  19. 19 Inkling 2 Small 1,653
  20. 20 Muse Wonder max 1,649
  21. 21 Grok 4.7 xhigh 1,641
  22. 22 MiMo-V2.6 Flash 1,638
  23. 23 Tencent Hy4 preview 1,632
  24. 24 Claude Fable 5 high 1,625
  25. 25 MiMo-V2.6 Pro 1,624
  26. 26 Qwen3.8-Flash-Next 1,624
  27. 27 GLM-5.3 max 1,622
  28. 28 Grok 4.6 high 1,618
  29. 29 GPT-5.6 Sol 1,615
  30. 30 GLM-5.3-Flash 1,609
  31. 31 Muse Ember max 1,609
  32. 32 GLM-5.2 max 1,608
  33. 33 Claude Haiku 5.5 high 1,594
  34. 34 Qwen 3.8 27B 1,590
  35. 35 Gemini 3.7 Flash high 1,588
  36. 36 MiniMax-M3.1-Flash 1,585
  37. 37 Gemini 3.8 Flash high 1,583
  38. 38 DeepSeek-V4-Flash high 1,582
  39. 39 DeepSeek-V4-Pro high 1,581
  40. 40 GPT-6 Luna max 1,577
  41. 41 Step 5 Preview high 1,573
  42. 42 Mistral Large 4 1,564
  43. 43 Claude Opus 4.7 high 1,557
  44. 44 Claude Opus 4.8 high 1,555
  45. 45 Grok 4.5 1,552
  46. 46 Claude Opus 4.6 high 1,547
  47. 47 Muse Spark 1.1 1,541
  48. 48 Claude Sonnet 5 high 1,539
  49. 49 Gemini 3.6 Flash high 1,538
  50. 50 Muse Spark 1.2 xhigh 1,534
  51. 51 Muse Spark Open max 1,526
  52. 52 Claude Sonnet 4.6 1,522
  53. 53 GPT-5.6 Terra 1,520
  54. 54 GPT-5.6 Luna 1,519
  55. 55 Seed2.1 Pro 1,519
  56. 56 Qwen3.7-Max 1,515
  57. 57 GPT-5.5 1,512
  58. 58 Kimi K2.6 1,509
  59. 59 GLM-5.1 1,508
  60. 60 Tencent Hy3 1,506
  61. 61 Gemini 3.5 Flash high 1,500
  62. 62 Claude Opus 4.5 high, 32k 1,493
  63. 63 MiniMax-M3 1,484
  64. 64 Qwen 3.6 Max (Preview) 1,482
  65. 65 MiMo-V2.5-Pro 1,479
  66. 66 Kimi K2.7 Code 1,474
  67. 67 GPT-5.4 1,462
  68. 68 Qwen 3.6 Plus 1,461
  69. 69 GLM-5 1,449
  70. 70 Gemini 3.1 Pro 1,446
  71. 71 Gemini 3 Pro 1,441
  72. 72 Gemini 3 Flash 1,440
  73. 73 MiMo-V2.5 1,438
  74. 74 Beam 1.0 1,437
  75. 75 Kimi K2.5 thinking 1,436
  76. 76 Gemini 3.5 Flash-Lite 1,435
  77. 77 GLM-4.7 1,435
  78. 78 MiMo-V2-Pro 1,434
  79. 79 GPT-5.2 1,420
  80. 80 GPT-5 medium 1,417
  81. 81 Inkling 1,414
  82. 82 Inkling-Small 1,408
  83. 83 GPT-5.3 Codex 1,406
  84. 84 Nemotron 3.5 Super VL nvfp4 1,401
  85. 85 Qwen3.5 397B-A17B 1,400
  86. 86 GLM-5V-Turbo 1,399
  87. 87 MiniMax-M2.7 1,398
  88. 88 GPT-5.4 Mini high 1,398
  89. 89 GPT-5.1 medium 1,397
  90. 90 Claude Sonnet 4.5 high, 32k 1,393
  91. 91 Claude Opus 4.1 1,389
  92. 92 MiniMax-M2.5 1,386
  93. 93 MiniMax-M2.1 1,384
  94. 94 Gemma 4 31B IT 1,370
  95. 95 Solar Pro 4 1,369
  96. 96 DeepSeek-V3.2 thinking 1,362
  97. 97 Qwen3.5-122B-A10B 1,360
  98. 98 Gemma 4 26B A4B 1,358
  99. 99 Tencent Hy3 preview 1,358
  100. 100 Grok 4.3 1,357
  101. 101 Qwen3.5-27B 1,357
  102. 102 Muse Glimmer 1,355
  103. 103 Laguna M.1 1,350
  104. 104 GLM-4.6 1,342
  105. 105 GPT-5.2 Codex 1,338
  106. 106 GPT-5.1-Codex 1,338
  107. 107 MiMo-V2-Flash non-thinking 1,331
  108. 108 Claude Haiku 4.5 1,330
  109. 109 Laguna XS.2 1,303
  110. 110 MiniMax-M2 1,299
  111. 111 Qwen3-Coder-480B-A35B 1,275
  112. 112 DeepSeek-V3.2-Exp 1,270
  113. 113 Mistral Medium 3.5 1,263
  114. 114 Gemini 3.1 Flash-Lite 1,256
  115. 115 Qwen3.5-35B-A3B 1,253
  116. 116 GPT-5.1-Codex-mini 1,245
  117. 117 Qwen3.5-Flash 1,244
  118. 118 Grok 4.1 Fast reasoning 1,242
  119. 119 Trinity Large Thinking 1,236
  120. 120 Mistral Large 3 1,233
  121. 121 Gemini 2.5 Pro Jun 2025 1,228
  122. 122 Grok 4.1 thinking 1,217
  123. 123 Devstral 2 1,196
  124. 124 Granite 4.1 8B 1,193
  125. 125 Mercury 2 1,171
  126. 126 grok-code-fast-1 1,164
  127. 127 Grok 4 Fast reasoning 1,162
  128. 128 Devstral Medium 1,072

Sources: Official Code Arena leaderboard and labs’ model cards

#EntryScoreSettingDateSource
1 Claude Fable 5.2 1,836 xhigh Nov 2026 Official leaderboard
2 Claude Opus 5.5 1,814 max Sep 2026 Official leaderboard
3 Claude Sonnet 5.5 1,773 xhigh Sep 2026 Official leaderboard
4 Gemini 4 Carbon 1,768 high Oct 2026 Official leaderboard
5 GPT-6 Astra 1,764 max Sep 2026 Official leaderboard
6 GPT-6.1 Sol 1,743 max Sep 2026 Official leaderboard
7 Claude Fable 5.1 1,742 max Sep 2026 Official leaderboard
8 Muse Wonder Full 1,723 max Oct 2026 Official leaderboard
9 Claude Opus 5 1,691 max Jul 2026 Official leaderboard
10 Kimi K3.1 Pro 1,688 max Oct 2026 Official leaderboard
11 Gemini 4 Argon 1,682 high Sep 2026 Official leaderboard
12 GPT-6.1 Terra 1,679 max Oct 2026 Official leaderboard
13 Qwen3.8 Max (0902) 1,676 Sep 2026 Official leaderboard
14 GPT-6 Sol 1,673 max Sep 2026 Official leaderboard
15 Qwen 3.8 Max 1,670 Aug 2026 Official leaderboard
16 DeepSeek V4.1 Flash 1,664 max Sep 2026 Official leaderboard
17 Kimi K3 1,658 max Jul 2026 Official leaderboard
18 Muse Spark 1.3 1,655 max Sep 2026 Official leaderboard
19 Muse Wonder 1,649 max Oct 2026 Official leaderboard
20 Inkling 2 Small 1,646 Nov 2026 Official leaderboard
21 Grok 4.7 1,641 xhigh Sep 2026 Official leaderboard
22 MiMo-V2.6 Flash 1,638 Sep 2026 Official leaderboard
23 Tencent Hy4 preview 1,632 Aug 2026 Official leaderboard
24 Claude Fable 5 1,625 high Jun 2026 Official leaderboard
25 MiMo-V2.6 Pro 1,624 Sep 2026 Official leaderboard
26 Qwen3.8-Flash-Next 1,624 Aug 2026 Official leaderboard
27 GLM-5.3 1,622 max Aug 2026 Official leaderboard
28 Grok 4.6 1,618 high Aug 2026 Official leaderboard
29 GPT-5.6 Sol 1,615 xHigh, Codex Jul 2026 Official leaderboard
30 U2 Flash 1,611 Nov 2026 Official leaderboard
31 GLM-5.3-Flash 1,609 Aug 2026 Official leaderboard
32 Muse Ember 1,609 max Nov 2026 Official leaderboard
33 GLM-5.2 1,608 max Jun 2026 Official leaderboard
34 Claude Haiku 5.5 1,594 high Oct 2026 Official leaderboard
35 Qwen 3.8 27B 1,590 Aug 2026 Official leaderboard
36 Gemini 3.7 Flash 1,588 high Aug 2026 Official leaderboard
37 MiniMax-M3.1-Flash 1,585 Oct 2026 Official leaderboard
38 Gemini 3.8 Flash 1,583 high Sep 2026 Official leaderboard
39 DeepSeek-V4-Flash 1,582 high Apr 2026 Official leaderboard
40 DeepSeek-V4-Pro 1,581 high Apr 2026 Official leaderboard
41 GPT-6 Luna 1,577 max Sep 2026 Official leaderboard
42 Step 5 Preview 1,573 high Sep 2026 Official leaderboard
43 Mistral Large 4 1,564 Oct 2026 Official leaderboard
44 Claude Opus 4.7 1,557 high Apr 2026 Official leaderboard
45 Claude Opus 4.8 1,555 high May 2026 Official leaderboard
46 Grok 4.5 1,552 Jul 2026 Official leaderboard
47 Claude Opus 4.6 1,547 high Feb 2026 Official leaderboard
48 Muse Spark 1.1 1,541 Jul 2026 Official leaderboard
49 Claude Sonnet 5 1,539 high Jun 2026 Official leaderboard
50 Gemini 3.6 Flash 1,538 high Jul 2026 Official leaderboard
51 Muse Spark 1.2 1,534 xhigh Aug 2026 Official leaderboard
52 Muse Spark Open 1,526 max Oct 2026 Official leaderboard
53 Claude Sonnet 4.6 1,522 Feb 2026 Official leaderboard
54 GPT-5.6 Terra 1,520 xHigh, Codex Jul 2026 Official leaderboard
55 GPT-5.6 Luna 1,519 xHigh, Codex Jul 2026 Official leaderboard
56 Seed2.1 Pro 1,519 Jun 2026 Official leaderboard
57 Qwen3.7-Max 1,515 May 2026 Official leaderboard
58 GPT-5.5 1,512 xHigh, Codex Apr 2026 Official leaderboard
59 Kimi K2.6 1,509 Apr 2026 Official leaderboard
60 GLM-5.1 1,508 Apr 2026 Official leaderboard
61 Tencent Hy3 1,506 Jul 2026 Official leaderboard
62 Gemini 3.5 Flash 1,500 high May 2026 Official leaderboard
63 Claude Opus 4.5 1,493 high, 32k Nov 2025 Official leaderboard
64 Spark-X2.6 1,485 Nov 2026 Official leaderboard
65 MiniMax-M3 1,484 Jun 2026 Official leaderboard
66 Qwen 3.6 Max (Preview) 1,482 Apr 2026 Official leaderboard
67 MiMo-V2.5-Pro 1,479 Apr 2026 Official leaderboard
68 Kimi K2.7 Code 1,474 Jun 2026 Official leaderboard
69 GPT-5.4 1,462 High, Codex Mar 2026 Official leaderboard
70 Qwen 3.6 Plus 1,461 Mar 2026 Official leaderboard
71 Gemini 3.1 Pro 1,446 Feb 2026 Official leaderboard
72 Gemini 3 Pro 1,441 Nov 2025 Official leaderboard
73 Gemini 3 Flash 1,440 Dec 2025 Official leaderboard
74 MiMo-V2.5 1,438 Apr 2026 Official leaderboard
75 Beam 1.0 1,437 Oct 2026 Official leaderboard
76 Kimi K2.5 1,436 thinking Jan 2026 Official leaderboard
77 Gemini 3.5 Flash-Lite 1,435 Jul 2026 Official leaderboard
78 GLM-5 1,435 Feb 2026 Official leaderboard
79 GLM-4.7 1,435 Dec 2025 Official leaderboard
80 MiMo-V2-Pro 1,434 Mar 2026 Official leaderboard
81 GPT-5.2 1,420 Dec 2025 Official leaderboard
82 GPT-5 1,417 medium Aug 2025 Official leaderboard
83 Inkling 1,414 Jul 2026 Official leaderboard
84 Inkling-Small 1,408 Jul 2026 Official leaderboard
85 GPT-5.3 Codex 1,406 Codex Feb 2026 Official leaderboard
86 Nemotron 3.5 Super VL 1,401 nvfp4 Oct 2026 Official leaderboard
87 Qwen3.5 397B-A17B 1,400 Feb 2026 Official leaderboard
88 GLM-5V-Turbo 1,399 Apr 2026 Official leaderboard
89 MiniMax-M2.7 1,398 Mar 2026 Official leaderboard
90 GPT-5.4 Mini 1,398 high Mar 2026 Official leaderboard
91 GPT-5.1 1,397 medium Nov 2025 Official leaderboard
92 Claude Sonnet 4.5 1,393 high, 32k Sep 2025 Official leaderboard
93 Claude Opus 4.1 1,389 Aug 2025 Official leaderboard
94 MiniMax-M2.5 1,386 Feb 2026 Official leaderboard
95 MiniMax-M2.1 1,384 Dec 2025 Official leaderboard
96 Grok 4.20 Beta (Reasoning, 0309) 1,375 Nov 2026 Official leaderboard
97 Gemma 4 31B IT 1,370 Apr 2026 Official leaderboard
98 Solar Pro 4 1,369 Aug 2026 Official leaderboard
99 DeepSeek-V3.2 1,362 thinking Dec 2025 Official leaderboard
100 Qwen3.5-122B-A10B 1,360 Feb 2026 Official leaderboard
101 Gemma 4 26B A4B 1,358 Apr 2026 Official leaderboard
102 Tencent Hy3 preview 1,358 Apr 2026 Official leaderboard
103 Grok 4.3 1,357 Apr 2026 Official leaderboard
104 Qwen3.5-27B 1,357 Feb 2026 Official leaderboard
105 Muse Glimmer 1,355 Aug 2026 Official leaderboard
106 Laguna M.1 1,350 Apr 2026 Official leaderboard
107 GLM-4.6 1,342 Sep 2025 Official leaderboard
108 GPT-5.2 Codex 1,338 Dec 2025 Official leaderboard
109 GPT-5.1-Codex 1,338 Nov 2025 Official leaderboard
110 MiMo-V2-Flash 1,331 non-thinking Dec 2025 Official leaderboard
111 Claude Haiku 4.5 1,330 Oct 2025 Official leaderboard
112 Kimi K2 Thinking Turbo 1,322 Nov 2026 Official leaderboard
113 Laguna XS.2 1,303 Apr 2026 Official leaderboard
114 MiniMax-M2 1,299 Oct 2025 Official leaderboard
115 Qwen3-Coder-480B-A35B 1,275 Jul 2025 Official leaderboard
116 DeepSeek-V3.2-Exp 1,270 Sep 2025 Official leaderboard
117 Mistral Medium 3.5 1,263 Apr 2026 Official leaderboard
118 KAT-Coder-Pro V1 1,258 Nov 2026 Official leaderboard
119 Gemini 3.1 Flash-Lite 1,256 Mar 2026 Official leaderboard
120 Qwen3.5-35B-A3B 1,253 Feb 2026 Official leaderboard
121 GPT-5.1-Codex-mini 1,245 Nov 2025 Official leaderboard
122 Qwen3.5-Flash 1,244 Feb 2026 Official leaderboard
123 Grok 4.1 Fast 1,242 reasoning Nov 2025 Official leaderboard
124 Trinity Large Thinking 1,236 Apr 2026 Official leaderboard
125 Mistral Large 3 1,233 Dec 2025 Official leaderboard
126 Gemini 2.5 Pro Jun 2025 1,228 Jun 2025 Official leaderboard
127 Grok 4.1 1,217 thinking Nov 2025 Official leaderboard
128 Devstral 2 1,196 Dec 2025 Official leaderboard
129 Granite 4.1 8B 1,193 Apr 2026 Official leaderboard
130 Mercury 2 1,171 Feb 2026 Official leaderboard
131 grok-code-fast-1 1,164 Aug 2025 Official leaderboard
132 Grok 4 Fast 1,162 reasoning Sep 2025 Official leaderboard
133 Devstral Medium 1,072 Jul 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Fable 5.2 1,836 xhigh Nov 2026 Official leaderboard
2 Claude Opus 5.5 1,814 max Sep 2026 Official leaderboard
3 Claude Sonnet 5.5 1,773 xhigh Sep 2026 Official leaderboard
4 Gemini 4 Carbon 1,768 high Oct 2026 Official leaderboard
5 GPT-6 Astra 1,764 max Sep 2026 Official leaderboard
6 GPT-6.1 Sol 1,743 max Sep 2026 Official leaderboard
7 Claude Fable 5.1 1,742 max Sep 2026 Official leaderboard
8 Muse Wonder Full 1,723 max Oct 2026 Official leaderboard
9 Claude Opus 5 1,691 max Jul 2026 Official leaderboard
10 Kimi K3.1 Pro 1,688 max Oct 2026 Official leaderboard
11 Gemini 4 Argon 1,682 high Sep 2026 Official leaderboard
12 GPT-6.1 Terra 1,679 max Oct 2026 Official leaderboard
13 Kimi K3 1,678 Jul 2026 Lab model card
14 Qwen3.8 Max (0902) 1,676 Sep 2026 Official leaderboard
15 GPT-6 Sol 1,673 max Sep 2026 Official leaderboard
16 Qwen 3.8 Max 1,670 Aug 2026 Official leaderboard
17 DeepSeek V4.1 Flash 1,664 max Sep 2026 Official leaderboard
18 Muse Spark 1.3 1,655 max Sep 2026 Official leaderboard
19 Inkling 2 Small 1,653 Nov 2026 Lab model card
20 Muse Wonder 1,649 max Oct 2026 Official leaderboard
21 Grok 4.7 1,641 xhigh Sep 2026 Official leaderboard
22 MiMo-V2.6 Flash 1,638 Sep 2026 Official leaderboard
23 Tencent Hy4 preview 1,632 Aug 2026 Official leaderboard
24 Claude Fable 5 1,625 high Jun 2026 Official leaderboard
25 MiMo-V2.6 Pro 1,624 Sep 2026 Official leaderboard
26 Qwen3.8-Flash-Next 1,624 Aug 2026 Official leaderboard
27 GLM-5.3 1,622 max Aug 2026 Official leaderboard
28 Grok 4.6 1,618 high Aug 2026 Official leaderboard
29 GPT-5.6 Sol 1,615 xHigh, Codex Jul 2026 Official leaderboard
30 GLM-5.3-Flash 1,609 Aug 2026 Official leaderboard
31 Muse Ember 1,609 max Nov 2026 Official leaderboard
32 GLM-5.2 1,608 max Jun 2026 Official leaderboard
33 Claude Haiku 5.5 1,594 high Oct 2026 Official leaderboard
34 Qwen 3.8 27B 1,590 Aug 2026 Official leaderboard
35 Gemini 3.7 Flash 1,588 high Aug 2026 Official leaderboard
36 MiniMax-M3.1-Flash 1,585 Oct 2026 Official leaderboard
37 Gemini 3.8 Flash 1,583 high Sep 2026 Official leaderboard
38 DeepSeek-V4-Flash 1,582 high Apr 2026 Official leaderboard
39 DeepSeek-V4-Pro 1,581 high Apr 2026 Official leaderboard
40 GPT-6 Luna 1,577 max Sep 2026 Official leaderboard
41 Step 5 Preview 1,573 high Sep 2026 Official leaderboard
42 Mistral Large 4 1,564 Oct 2026 Official leaderboard
43 Claude Opus 4.7 1,557 high Apr 2026 Official leaderboard
44 Claude Opus 4.8 1,555 high May 2026 Official leaderboard
45 Grok 4.5 1,552 Jul 2026 Official leaderboard
46 Claude Opus 4.6 1,547 high Feb 2026 Official leaderboard
47 Muse Spark 1.1 1,541 Jul 2026 Official leaderboard
48 Claude Sonnet 5 1,539 high Jun 2026 Official leaderboard
49 Gemini 3.6 Flash 1,538 high Jul 2026 Official leaderboard
50 Muse Spark 1.2 1,534 xhigh Aug 2026 Official leaderboard
51 Muse Spark Open 1,526 max Oct 2026 Official leaderboard
52 Claude Sonnet 4.6 1,522 Feb 2026 Official leaderboard
53 GPT-5.6 Terra 1,520 xHigh, Codex Jul 2026 Official leaderboard
54 GPT-5.6 Luna 1,519 xHigh, Codex Jul 2026 Official leaderboard
55 Seed2.1 Pro 1,519 Jun 2026 Official leaderboard
56 Qwen3.7-Max 1,515 May 2026 Official leaderboard
57 GPT-5.5 1,512 xHigh, Codex Apr 2026 Official leaderboard
58 Kimi K2.6 1,509 Apr 2026 Official leaderboard
59 GLM-5.1 1,508 Apr 2026 Official leaderboard
60 Tencent Hy3 1,506 Jul 2026 Official leaderboard
61 Gemini 3.5 Flash 1,500 high May 2026 Official leaderboard
62 Claude Opus 4.5 1,493 high, 32k Nov 2025 Official leaderboard
63 MiniMax-M3 1,484 Jun 2026 Official leaderboard
64 Qwen 3.6 Max (Preview) 1,482 Apr 2026 Official leaderboard
65 MiMo-V2.5-Pro 1,479 Apr 2026 Official leaderboard
66 Kimi K2.7 Code 1,474 Jun 2026 Official leaderboard
67 GPT-5.4 1,462 High, Codex Mar 2026 Official leaderboard
68 Qwen 3.6 Plus 1,461 Mar 2026 Official leaderboard
69 GLM-5 1,449 Feb 2026 Lab model card
70 Gemini 3.1 Pro 1,446 Feb 2026 Official leaderboard
71 Gemini 3 Pro 1,441 Nov 2025 Official leaderboard
72 Gemini 3 Flash 1,440 Dec 2025 Official leaderboard
73 MiMo-V2.5 1,438 Apr 2026 Official leaderboard
74 Beam 1.0 1,437 Oct 2026 Official leaderboard
75 Kimi K2.5 1,436 thinking Jan 2026 Official leaderboard
76 Gemini 3.5 Flash-Lite 1,435 Jul 2026 Official leaderboard
77 GLM-4.7 1,435 Dec 2025 Official leaderboard
78 MiMo-V2-Pro 1,434 Mar 2026 Official leaderboard
79 GPT-5.2 1,420 Dec 2025 Official leaderboard
80 GPT-5 1,417 medium Aug 2025 Official leaderboard
81 Inkling 1,414 Jul 2026 Official leaderboard
82 Inkling-Small 1,408 Jul 2026 Official leaderboard
83 GPT-5.3 Codex 1,406 Codex Feb 2026 Official leaderboard
84 Nemotron 3.5 Super VL 1,401 nvfp4 Oct 2026 Official leaderboard
85 Qwen3.5 397B-A17B 1,400 Feb 2026 Official leaderboard
86 GLM-5V-Turbo 1,399 Apr 2026 Official leaderboard
87 MiniMax-M2.7 1,398 Mar 2026 Official leaderboard
88 GPT-5.4 Mini 1,398 high Mar 2026 Official leaderboard
89 GPT-5.1 1,397 medium Nov 2025 Official leaderboard
90 Claude Sonnet 4.5 1,393 high, 32k Sep 2025 Official leaderboard
91 Claude Opus 4.1 1,389 Aug 2025 Official leaderboard
92 MiniMax-M2.5 1,386 Feb 2026 Official leaderboard
93 MiniMax-M2.1 1,384 Dec 2025 Official leaderboard
94 Gemma 4 31B IT 1,370 Apr 2026 Official leaderboard
95 Solar Pro 4 1,369 Aug 2026 Official leaderboard
96 DeepSeek-V3.2 1,362 thinking Dec 2025 Official leaderboard
97 Qwen3.5-122B-A10B 1,360 Feb 2026 Official leaderboard
98 Gemma 4 26B A4B 1,358 Apr 2026 Official leaderboard
99 Tencent Hy3 preview 1,358 Apr 2026 Official leaderboard
100 Grok 4.3 1,357 Apr 2026 Official leaderboard
101 Qwen3.5-27B 1,357 Feb 2026 Official leaderboard
102 Muse Glimmer 1,355 Aug 2026 Official leaderboard
103 Laguna M.1 1,350 Apr 2026 Official leaderboard
104 GLM-4.6 1,342 Sep 2025 Official leaderboard
105 GPT-5.2 Codex 1,338 Dec 2025 Official leaderboard
106 GPT-5.1-Codex 1,338 Nov 2025 Official leaderboard
107 MiMo-V2-Flash 1,331 non-thinking Dec 2025 Official leaderboard
108 Claude Haiku 4.5 1,330 Oct 2025 Official leaderboard
109 Laguna XS.2 1,303 Apr 2026 Official leaderboard
110 MiniMax-M2 1,299 Oct 2025 Official leaderboard
111 Qwen3-Coder-480B-A35B 1,275 Jul 2025 Official leaderboard
112 DeepSeek-V3.2-Exp 1,270 Sep 2025 Official leaderboard
113 Mistral Medium 3.5 1,263 Apr 2026 Official leaderboard
114 Gemini 3.1 Flash-Lite 1,256 Mar 2026 Official leaderboard
115 Qwen3.5-35B-A3B 1,253 Feb 2026 Official leaderboard
116 GPT-5.1-Codex-mini 1,245 Nov 2025 Official leaderboard
117 Qwen3.5-Flash 1,244 Feb 2026 Official leaderboard
118 Grok 4.1 Fast 1,242 reasoning Nov 2025 Official leaderboard
119 Trinity Large Thinking 1,236 Apr 2026 Official leaderboard
120 Mistral Large 3 1,233 Dec 2025 Official leaderboard
121 Gemini 2.5 Pro Jun 2025 1,228 Jun 2025 Official leaderboard
122 Grok 4.1 1,217 thinking Nov 2025 Official leaderboard
123 Devstral 2 1,196 Dec 2025 Official leaderboard
124 Granite 4.1 8B 1,193 Apr 2026 Official leaderboard
125 Mercury 2 1,171 Feb 2026 Official leaderboard
126 grok-code-fast-1 1,164 Aug 2025 Official leaderboard
127 Grok 4 Fast 1,162 reasoning Sep 2025 Official leaderboard
128 Devstral Medium 1,072 Jul 2025 Official leaderboard

Labs report scores for 4 models in their cards. Switch the scores above to rank them together.

More effort adds a median 35 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
Claude Sonnet 5.5
high xhigh
1,773
Claude Opus 5
high
1,691
Muse Spark 1.3
xhigh
1,655
DeepSeek-V4-Pro
high high
1,581
Claude Opus 4.7
high
1,557
Claude Opus 4.8
high
1,555
Claude Opus 4.6
high
1,547
Gemini 3.5 Flash
medium
1,500
Claude Opus 4.5
high, 32k
1,493
Gemini 3 Flash
minimal
1,440
Kimi K2.5
instant
1,436
GPT-5.1
medium
1,397

Source: Official Code Arena leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 xhigh 1,836
Claude Opus 5.5 max 1,814
Claude Sonnet 5.5 xhigh 1,773
Gemini 4 Carbon high 1,768
GPT-6 Astra max 1,764
GPT-6.1 Sol max 1,743
Claude Fable 5.1 max 1,742
Muse Wonder Full max 1,723
Claude Sonnet 5.5 high 1,715
Claude Opus 5 max 1,691
Kimi K3.1 Pro max 1,688
Gemini 4 Argon high 1,682
GPT-6.1 Terra max 1,679
Qwen3.8 Max (0902) 1,676
GPT-6 Sol max 1,673
Qwen 3.8 Max 1,670
DeepSeek V4.1 Flash max 1,664
Kimi K3 max 1,658
Claude Opus 5 high 1,657
Muse Spark 1.3 max 1,655
Muse Wonder max 1,649
Inkling 2 Small 1,646
Grok 4.7 xhigh 1,641
MiMo-V2.6 Flash 1,638
Tencent Hy4 preview 1,632
Claude Fable 5 high 1,625
MiMo-V2.6 Pro 1,624
Qwen3.8-Flash-Next 1,624
GLM-5.3 max 1,622
DeepSeek V4.1 Flash max 1,621
Muse Spark 1.3 xhigh 1,620
Grok 4.6 high 1,618
GPT-5.6 Sol xHigh, Codex 1,615
GLM-5.3-Flash 1,609
Muse Ember max 1,609
GLM-5.2 max 1,608
Claude Haiku 5.5 high 1,594
Qwen 3.8 27B 1,590
Gemini 3.7 Flash high 1,588
MiniMax-M3.1-Flash 1,585
Gemini 3.8 Flash high 1,583
DeepSeek-V4-Flash high 1,582
DeepSeek-V4-Pro high 1,581
GPT-6 Luna max 1,577
Step 5 Preview high 1,573
Mistral Large 4 1,564
Claude Opus 4.7 high 1,557
Claude Opus 4.8 high 1,555
Claude Opus 4.7 1,555
Grok 4.5 1,552
Claude Opus 4.6 high 1,547
Muse Spark 1.1 1,541
Claude Sonnet 5 high 1,539
Gemini 3.6 Flash high 1,538
Claude Opus 4.6 1,537
Claude Opus 4.8 1,536
Muse Spark 1.2 xhigh 1,534
Muse Spark Open max 1,526
Claude Sonnet 4.6 1,522
GPT-5.6 Terra xHigh, Codex 1,520
GPT-5.6 Luna xHigh, Codex 1,519
Seed2.1 Pro 1,519
Qwen3.7-Max 1,515
GPT-5.5 xHigh, Codex 1,512
Kimi K2.6 1,509
GLM-5.1 1,508
Tencent Hy3 1,506
Gemini 3.5 Flash high 1,500
Claude Opus 4.5 high, 32k 1,493
Gemini 3.5 Flash medium 1,490
GPT-5.5 High, Codex 1,487
MiniMax-M3 1,484
Qwen 3.6 Max (Preview) 1,482
MiMo-V2.5-Pro 1,479
Kimi K2.7 Code 1,474
Claude Opus 4.5 1,469
DeepSeek-V4-Pro high 1,463
GPT-5.4 High, Codex 1,462
Qwen 3.6 Plus 1,461
GPT-5.5 Codex 1,455
DeepSeek-V4-Pro 1,448
Gemini 3.1 Pro 1,446
GPT-5.4 Medium, Codex 1,442
Gemini 3 Pro 1,441
Gemini 3 Flash 1,440
MiMo-V2.5 1,438
Beam 1.0 1,437
Kimi K2.5 thinking 1,436
Gemini 3.5 Flash-Lite 1,435
GLM-5 1,435
GLM-4.7 1,435
MiMo-V2-Pro 1,434
DeepSeek-V4-Flash high 1,431
GPT-5.2 1,420
GPT-5 medium 1,417
Inkling 1,414
Inkling-Small 1,408
GPT-5.3 Codex Codex 1,406
Kimi K2.5 instant 1,405
Nemotron 3.5 Super VL nvfp4 1,401
Qwen3.5 397B-A17B 1,400
GPT-5.4 1,399
GLM-5V-Turbo 1,399
MiniMax-M2.7 1,398
GPT-5.4 Mini high 1,398
GPT-5.1 medium 1,397
Claude Sonnet 4.5 high, 32k 1,393
Claude Opus 4.1 1,389
MiniMax-M2.5 1,386
Claude Sonnet 4.5 1,385
MiniMax-M2.1 1,384
Gemini 3 Flash minimal 1,384
GPT-5.3 Codex Codex 1,371
Gemma 4 31B IT 1,370
Solar Pro 4 1,369
DeepSeek-V3.2 thinking 1,362
Qwen3.5-122B-A10B 1,360
Gemma 4 26B A4B 1,358
Tencent Hy3 preview 1,358
Grok 4.3 1,357
Qwen3.5-27B 1,357
Muse Glimmer 1,355
Laguna M.1 1,350
GLM-4.6 1,342
GPT-5.1 1,342
GPT-5.2 Codex 1,338
GPT-5.1-Codex 1,338
MiMo-V2-Flash non-thinking 1,331
Claude Haiku 4.5 1,330
DeepSeek-V3.2 1,326
Laguna XS.2 1,303
MiniMax-M2 1,299
MiMo-V2-Flash thinking 1,292
Qwen3-Coder-480B-A35B 1,275
DeepSeek-V3.2-Exp 1,270
Mistral Medium 3.5 1,263
Gemini 3.1 Flash-Lite 1,256
Qwen3.5-35B-A3B 1,253
GPT-5.1-Codex-mini 1,245
Qwen3.5-Flash 1,244
Grok 4.1 Fast reasoning 1,242
Trinity Large Thinking 1,236
Mistral Large 3 1,233
Gemini 2.5 Pro Jun 2025 1,228
Grok 4.1 thinking 1,217
Devstral 2 1,196
Granite 4.1 8B 1,193
Mercury 2 1,171
grok-code-fast-1 1,164
Grok 4 Fast reasoning 1,162
Devstral Medium 1,072

13 models have held number one on the official leaderboard since its release in November 2025 13 models have held number one with claims included since its release in November 2025

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.

OpenAI Google DeepMind Anthropic DeepSeek Moonshot AI Z.ai
Own leaderboard 13 leaders
Claude Fable 5.2 1,836
Lab claims 2 leaders
Kimi K3 1,678

Sources: Official Code Arena leaderboard and labs’ model cards

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 Reported by the lab in its model card
1,000 1,200 1,400 1,600 1,800 2,000 2025 2026 2027 1,000 1,200 1,400 1,600 1,800 2,000 2025 2026 2027

Sources: Official Code Arena leaderboard and labs’ model cards

2 labs put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Google DeepMind
Thinking Machines
DateLaunch postScoreAs the post writes it
Nov 5, 2026 Inkling 2 Small 1653 Code Arena, 1653 Elo
Aug 13, 2026 Gemini 3.7 Flash 1588 Code Arena (Web development, Elo)

3 system cards report it, from 3 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Aug 13, 2026 Gemini 3.7 Flash model card 1588 Code Arena, web development, Elo (results table, p.5) Capabilities
Jul 27, 2026 Kimi K3 technical report 1678 WebDev Arena (Elo), rank 1 of 99 (Table 5). Mapped to the current Arena WebDev board Other
Feb 17, 2026 GLM-5 technical report 1449 LMArena Code Arena (Agentic WebDev) score, rank 6 (Figure 3 screenshot) Capabilities

4 lab claims next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model Lab says Board says Gap What differs
Kimi K3 1,678 WebDev Arena (Elo), rank 1 of 99 (Table 5). Mapped to the current Arena WebDev board 1,658 Leaderboard, max +20 Moonshot quotes Arena's July 23 rating, while the board reflects votes collected through November.
GLM-5 1,449 LMArena Code Arena (Agentic WebDev) score, rank 6 (Figure 3 screenshot) 1,435 Leaderboard +14 The lab reproduces an early Code Arena snapshot, while the board uses a later rating from additional votes.
Inkling 2 Small 1,653 Code Arena, 1653 Elo 1,646 Leaderboard +7
Gemini 3.7 Flash 1,588 Code Arena (Web development, Elo) 1,588 Leaderboard, high Same Google quotes an August Arena rating as an integer, while the board supplies a later rating with decimal precision.

Versions

Replaces WebDev Arena

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