Benchmark Almanac

SimpleBench

Claude Fable 5.2 94.9% leads since November 2026
Claude Fable 5.2 94.9% leads since November 2026
Not in model cards

Models answer over 200 multiple-choice trick questions on spatial, temporal, and social reasoning that ordinary people solve. Scored by average accuracy over five runs.

The human baseline from nine non-expert people is 83.7 percent. A small open-ended variant lists three models.

Released
October 2024
Kind
Abstract puzzles and logic
Level
Everyday
Scored
Checked automatically
Answers
Multiple choice
Tasks
Public, plus a private test set
Strongest model tested
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
110
Entries on its own leaderboard, 106 models we track

Claude Fable 5.2 leads with 94.9%

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. 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 94.9%, 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.

110 entries. Scroll sideways for the rest.

  1. 94.9 Claude Fable 5.2
  2. 92.6 Gemini 4 Carbon
  3. 91.7 Gemini 4 Argon
  4. 88.4 Claude Opus 5.5
  5. 86.6 Claude Fable 5.1
  6. 86.5 GPT-6 Astra Pro
  7. 83.6 GPT-6 Astra
  8. 82.9 GPT-6.1 Sol
  9. 82.4 Gemini 3.8 Flash
  10. 81.9 Claude Fable 5
  11. 81.8 Muse Spark 1.3
  12. 80.6 Claude Opus 5
  13. 79.6 Gemini 3.1 Pro
  14. 76.9 GPT-5.5 Pro
  15. 76.7 Gemini 3.5 Flash
  16. 76.4 Gemini 3 Pro
  17. 75.9 Grok 4.6
  18. 75.9 Claude Sonnet 5.5
  19. 74.5 Muse Spark 1.2
  20. 73.1 GPT-6 Sol
  21. 72.7 Muse Ember
  22. 72.2 GPT-6.1 Terra
  23. 71.7 GPT-5.6 Sol Pro
  24. 70.4 Qwen3.7-Max
  25. 70.0 Grok 4.5
  26. 69.0 GPT-5.5
  27. 67.6 Claude Opus 4.6
  28. 66.7 DeepSeek V4.1 Flash
  29. 66.2 GLM-5.3
  30. 64.8 Claude Opus 4.8
  31. 64.8 GPT-5.6 Sol
  32. 63.0 Qwen 3.6 Max (Preview)
  33. 62.5 Qwen 3.8 Max
  34. 62.4 Gemini 2.5 Pro
  35. 62.0 Claude Opus 4.5
  36. 61.7 Claude Opus 4.7
  37. 61.6 GPT-5 Pro
  38. 61.1 Gemini 3 Flash
  39. 61.1 DeepSeek-V4-Flash
  40. 60.7 Kimi K3
  41. 60.6 Claude Sonnet 5
  42. 60.5 Grok 4
  43. 60.2 Qwen 3.8 27B
  44. 60.0 Claude Opus 4.1
  45. 59.2 Claude Haiku 5.5
  46. 58.8 Claude Opus 4
  47. 58.8 GLM-5.2
  48. 57.9 Kimi K2.7 Code
  49. 57.4 GPT-5.2 Pro
  50. 56.7 GPT-5
  51. 56.0 Grok 4.1 Fast
  52. 55.1 GLM-5.1
  53. 55.1 Ox Alpha
  54. 54.3 Claude Sonnet 4.5
  55. 53.2 GPT-5.1
  56. 53.2 GLM-5
  57. 53.1 o3
  58. 52.6 DeepSeek-V3.2-Speciale
  59. 51.6 Gemini 2.5 Pro
  60. 50.9 DeepSeek-V4-Pro
  61. 50.0 Inkling
  62. 48.9 GPT-5.6 Terra
  63. 47.7 GLM-4.7
  64. 46.8 Kimi K2.5
  65. 46.8 GPT-5.6 Luna
  66. 46.4 Claude 3.7 Sonnet
  67. 46.3 DeepSeek V4 Flash (preview)
  68. 45.8 GPT-5.2
  69. 45.8 MiniMax-M3
  70. 45.5 Claude Sonnet 4
  71. 41.7 o1-preview
  72. 41.7 Nemotron 3 Ultra
  73. 41.4 Claude 3.5 Sonnet (October 2024)
  74. 41.2 Gemini 2.5 Flash
  75. 40.8 DeepSeek-R1
  76. 40.1 o1
  77. 40.0 DeepSeek-V3.1
  78. 39.6 Kimi K2 Thinking
  79. 38.7 o4-mini
  80. 36.1 Grok 3
  81. 35.2 Qwen 3.6 Flash
  82. 34.7 MiniMax-M2.1
  83. 34.5 GPT-4.5
  84. 31.1 Gemini-exp-1206
  85. 31.0 Qwen3-235B-A22B
  86. 30.9 DeepSeek-R1
  87. 30.7 Gemini 2.0 Flash Thinking
  88. 27.7 Llama 4 Maverick
  89. 27.5 Claude 3.5 Sonnet
  90. 27.2 DeepSeek-V3
  91. 27.1 Gemini 1.5 Pro
  92. 27.0 GPT-4.1
  93. 26.3 Kimi K2
  94. 25.1 GPT-4 Turbo
  95. 25.0 MiniMax-M2
  96. 23.5 Claude 3 Opus
  97. 23.0 Llama 3.1-405B
  98. 22.8 o3-mini
  99. 22.7 Grok-2
  100. 22.5 Mistral Large v2
  101. 22.1 gpt-oss-120b
  102. 20.4 Mistral Large 3
  103. 19.9 Llama 3.3 70B Instruct
  104. 18.9 Gemini 2.0 Flash
  105. 18.9 DeepSeek-V3
  106. 18.1 o1-mini
  107. 17.8 GPT-4o
  108. 17.4 Command R+
  109. 10.7 GPT-4o mini
  110. 8.00 GPT-3.5 Turbo
  1. 1 Claude Fable 5.2 94.9%
  2. 2 Gemini 4 Carbon 92.6%
  3. 3 Gemini 4 Argon 91.7%
  4. 4 Claude Opus 5.5 88.4%
  5. 5 Claude Fable 5.1 86.6%
  6. 6 GPT-6 Astra Pro 86.5%
  7. 7 GPT-6 Astra 83.6%
  8. 8 GPT-6.1 Sol 82.9%
  9. 9 Gemini 3.8 Flash 82.4%
  10. 10 Claude Fable 5 81.9%
  11. 11 Muse Spark 1.3 81.8%
  12. 12 Claude Opus 5 80.6%
  13. 13 Gemini 3.1 Pro 79.6%
  14. 14 GPT-5.5 Pro 76.9%
  15. 15 Gemini 3.5 Flash 76.7%
  16. 16 Gemini 3 Pro 76.4%
  17. 17 Grok 4.6 75.9%
  18. 18 Claude Sonnet 5.5 75.9%
  19. 19 Muse Spark 1.2 74.5%
  20. 20 GPT-6 Sol 73.1%
  21. 21 Muse Ember 72.7%
  22. 22 GPT-6.1 Terra 72.2%
  23. 23 GPT-5.6 Sol Pro xhigh 71.7%
  24. 24 Qwen3.7-Max 70.4%
  25. 25 Grok 4.5 70%
  26. 26 GPT-5.5 69%
  27. 27 Claude Opus 4.6 67.6%
  28. 28 DeepSeek V4.1 Flash 66.7%
  29. 29 GLM-5.3 66.2%
  30. 30 Claude Opus 4.8 64.8%
  31. 31 GPT-5.6 Sol xhigh 64.8%
  32. 32 Qwen 3.6 Max (Preview) 63%
  33. 33 Qwen 3.8 Max 62.5%
  34. 34 Gemini 2.5 Pro Jun 2025 62.4%
  35. 35 Claude Opus 4.5 62%
  36. 36 Claude Opus 4.7 61.7%
  37. 37 GPT-5 Pro 61.6%
  38. 38 Gemini 3 Flash 61.1%
  39. 39 DeepSeek-V4-Flash 61.1%
  40. 40 Kimi K3 max 60.7%
  41. 41 Claude Sonnet 5 60.6%
  42. 42 Grok 4 60.5%
  43. 43 Qwen 3.8 27B 60.2%
  44. 44 Claude Opus 4.1 60%
  45. 45 Claude Haiku 5.5 59.2%
  46. 46 Claude Opus 4 58.8%
  47. 47 GLM-5.2 58.8%
  48. 48 Kimi K2.7 Code 57.9%
  49. 49 GPT-5.2 Pro xhigh 57.4%
  50. 50 GPT-5 high 56.7%
  51. 51 Grok 4.1 Fast 56%
  52. 52 GLM-5.1 55.1%
  53. 53 Ox Alpha 55.1%
  54. 54 Claude Sonnet 4.5 54.3%
  55. 55 GPT-5.1 high 53.2%
  56. 56 GLM-5 53.2%
  57. 57 o3 high 53.1%
  58. 58 DeepSeek-V3.2-Speciale 52.6%
  59. 59 Gemini 2.5 Pro Mar 2025 51.6%
  60. 60 DeepSeek-V4-Pro 50.9%
  61. 61 Inkling 50%
  62. 62 GPT-5.6 Terra xhigh 48.9%
  63. 63 GLM-4.7 47.7%
  64. 64 Kimi K2.5 46.8%
  65. 65 GPT-5.6 Luna xhigh 46.8%
  66. 66 Claude 3.7 Sonnet thinking 46.4%
  67. 67 DeepSeek V4 Flash (preview) 46.3%
  68. 68 GPT-5.2 high 45.8%
  69. 69 MiniMax-M3 45.8%
  70. 70 Claude Sonnet 4 thinking 45.5%
  71. 71 o1-preview 41.7%
  72. 72 Nemotron 3 Ultra 41.7%
  73. 73 Claude 3.5 Sonnet (October 2024) 41.4%
  74. 74 Gemini 2.5 Flash Sep 2025 41.2%
  75. 75 DeepSeek-R1 May 2025 40.8%
  76. 76 o1 high 40.1%
  77. 77 DeepSeek-V3.1 40%
  78. 78 Kimi K2 Thinking 39.6%
  79. 79 o4-mini high 38.7%
  80. 80 Grok 3 36.1%
  81. 81 Qwen 3.6 Flash 35.2%
  82. 82 MiniMax-M2.1 34.7%
  83. 83 GPT-4.5 34.5%
  84. 84 Gemini-exp-1206 31.1%
  85. 85 Qwen3-235B-A22B 31%
  86. 86 DeepSeek-R1 30.9%
  87. 87 Gemini 2.0 Flash Thinking Jan 2025 30.7%
  88. 88 Llama 4 Maverick 27.7%
  89. 89 Claude 3.5 Sonnet 27.5%
  90. 90 DeepSeek-V3 Mar 2025 27.2%
  91. 91 Gemini 1.5 Pro Sep 2024 27.1%
  92. 92 GPT-4.1 27%
  93. 93 Kimi K2 Jul 2025 26.3%
  94. 94 GPT-4 Turbo Apr 2024 25.1%
  95. 95 MiniMax-M2 25%
  96. 96 Claude 3 Opus 23.5%
  97. 97 Llama 3.1-405B 23%
  98. 98 o3-mini high 22.8%
  99. 99 Grok-2 Aug 2024 22.7%
  100. 100 Mistral Large v2 22.5%
  101. 101 gpt-oss-120b 22.1%
  102. 102 Mistral Large 3 20.4%
  103. 103 Llama 3.3 70B Instruct 19.9%
  104. 104 Gemini 2.0 Flash Dec 2024 18.9%
  105. 105 DeepSeek-V3 18.9%
  106. 106 o1-mini 18.1%
  107. 107 GPT-4o Aug 2024 17.8%
  108. 108 Command R+ 17.4%
  109. 109 GPT-4o mini 10.7%
  110. 110 GPT-3.5 Turbo Jun 2023 8.00%

Source: Official SimpleBench leaderboard

107 models. Scroll sideways for the rest.

  1. 94.9 Claude Fable 5.2
  2. 92.6 Gemini 4 Carbon
  3. 91.7 Gemini 4 Argon
  4. 88.4 Claude Opus 5.5
  5. 86.6 Claude Fable 5.1
  6. 86.5 GPT-6 Astra Pro
  7. 83.6 GPT-6 Astra
  8. 82.9 GPT-6.1 Sol
  9. 82.4 Gemini 3.8 Flash
  10. 81.9 Claude Fable 5
  11. 81.8 Muse Spark 1.3
  12. 80.6 Claude Opus 5
  13. 79.6 Gemini 3.1 Pro
  14. 76.9 GPT-5.5 Pro
  15. 76.7 Gemini 3.5 Flash
  16. 76.4 Gemini 3 Pro
  17. 75.9 Grok 4.6
  18. 75.9 Claude Sonnet 5.5
  19. 74.5 Muse Spark 1.2
  20. 73.1 GPT-6 Sol
  21. 72.7 Muse Ember
  22. 72.2 GPT-6.1 Terra
  23. 71.7 GPT-5.6 Sol Pro
  24. 71.7 GPT-5.6 Sol
  25. 70.4 Qwen3.7-Max
  26. 70.0 Grok 4.5
  27. 69.0 GPT-5.5
  28. 67.6 Claude Opus 4.6
  29. 66.7 DeepSeek V4.1 Flash
  30. 66.2 GLM-5.3
  31. 64.8 Claude Opus 4.8
  32. 63.0 Qwen 3.6 Max (Preview)
  33. 62.5 Qwen 3.8 Max
  34. 62.4 Gemini 2.5 Pro
  35. 62.0 Claude Opus 4.5
  36. 61.7 Claude Opus 4.7
  37. 61.6 GPT-5 Pro
  38. 61.1 Gemini 3 Flash
  39. 61.1 DeepSeek-V4-Flash
  40. 61.1 DeepSeek V4 Flash 0731
  41. 60.7 Kimi K3
  42. 60.6 Claude Sonnet 5
  43. 60.5 Grok 4
  44. 60.2 Qwen 3.8 27B
  45. 60.0 Claude Opus 4.1
  46. 59.2 Claude Haiku 5.5
  47. 58.8 Claude Opus 4
  48. 58.8 GLM-5.2
  49. 57.9 Kimi K2.7 Code
  50. 57.4 GPT-5.2 Pro
  51. 56.7 GPT-5
  52. 56.0 Grok 4.1 Fast
  53. 55.1 GLM-5.1
  54. 54.3 Claude Sonnet 4.5
  55. 53.2 GPT-5.1
  56. 53.2 GLM-5
  57. 53.1 o3
  58. 52.6 DeepSeek-V3.2-Speciale
  59. 51.6 Gemini 2.5 Pro
  60. 50.9 DeepSeek-V4-Pro
  61. 50.0 Inkling
  62. 48.9 GPT-5.6 Terra
  63. 47.7 GLM-4.7
  64. 46.8 Kimi K2.5
  65. 46.8 GPT-5.6 Luna
  66. 46.4 Claude 3.7 Sonnet
  67. 45.8 GPT-5.2
  68. 45.8 MiniMax-M3
  69. 45.5 Claude Sonnet 4
  70. 41.7 o1-preview
  71. 41.7 Nemotron 3 Ultra
  72. 41.4 Claude 3.5 Sonnet (October 2024)
  73. 41.2 Gemini 2.5 Flash
  74. 40.8 DeepSeek-R1
  75. 40.1 o1
  76. 40.0 DeepSeek-V3.1
  77. 39.6 Kimi K2 Thinking
  78. 38.7 o4-mini
  79. 36.1 Grok 3
  80. 35.2 Qwen 3.6 Flash
  81. 34.7 MiniMax-M2.1
  82. 34.5 GPT-4.5
  83. 31.0 Qwen3-235B-A22B
  84. 30.9 DeepSeek-R1
  85. 30.7 Gemini 2.0 Flash Thinking
  86. 27.7 Llama 4 Maverick
  87. 27.5 Claude 3.5 Sonnet
  88. 27.2 DeepSeek-V3
  89. 27.1 Gemini 1.5 Pro
  90. 27.0 GPT-4.1
  91. 26.3 Kimi K2
  92. 25.1 GPT-4 Turbo
  93. 25.0 MiniMax-M2
  94. 23.5 Claude 3 Opus
  95. 23.0 Llama 3.1-405B
  96. 22.8 o3-mini
  97. 22.7 Grok-2
  98. 22.1 gpt-oss-120b
  99. 20.4 Mistral Large 3
  100. 19.9 Llama 3.3 70B Instruct
  101. 18.9 Gemini 2.0 Flash
  102. 18.9 DeepSeek-V3
  103. 18.1 o1-mini
  104. 17.8 GPT-4o
  105. 17.4 Command R+
  106. 10.7 GPT-4o mini
  107. 8.00 GPT-3.5 Turbo
  1. 1 Claude Fable 5.2 94.9%
  2. 2 Gemini 4 Carbon 92.6%
  3. 3 Gemini 4 Argon 91.7%
  4. 4 Claude Opus 5.5 88.4%
  5. 5 Claude Fable 5.1 86.6%
  6. 6 GPT-6 Astra Pro 86.5%
  7. 7 GPT-6 Astra 83.6%
  8. 8 GPT-6.1 Sol 82.9%
  9. 9 Gemini 3.8 Flash 82.4%
  10. 10 Claude Fable 5 81.9%
  11. 11 Muse Spark 1.3 81.8%
  12. 12 Claude Opus 5 80.6%
  13. 13 Gemini 3.1 Pro 79.6%
  14. 14 GPT-5.5 Pro 76.9%
  15. 15 Gemini 3.5 Flash 76.7%
  16. 16 Gemini 3 Pro 76.4%
  17. 17 Grok 4.6 75.9%
  18. 18 Claude Sonnet 5.5 75.9%
  19. 19 Muse Spark 1.2 74.5%
  20. 20 GPT-6 Sol 73.1%
  21. 21 Muse Ember 72.7%
  22. 22 GPT-6.1 Terra 72.2%
  23. 23 GPT-5.6 Sol Pro xhigh 71.7%
  24. 24 GPT-5.6 Sol pro, xhigh 71.7%
  25. 25 Qwen3.7-Max 70.4%
  26. 26 Grok 4.5 70%
  27. 27 GPT-5.5 69%
  28. 28 Claude Opus 4.6 67.6%
  29. 29 DeepSeek V4.1 Flash 66.7%
  30. 30 GLM-5.3 66.2%
  31. 31 Claude Opus 4.8 64.8%
  32. 32 Qwen 3.6 Max (Preview) 63%
  33. 33 Qwen 3.8 Max 62.5%
  34. 34 Gemini 2.5 Pro Jun 2025 62.4%
  35. 35 Claude Opus 4.5 62%
  36. 36 Claude Opus 4.7 61.7%
  37. 37 GPT-5 Pro 61.6%
  38. 38 Gemini 3 Flash 61.1%
  39. 39 DeepSeek-V4-Flash 61.1%
  40. 40 DeepSeek V4 Flash 0731 61.1%
  41. 41 Kimi K3 max 60.7%
  42. 42 Claude Sonnet 5 60.6%
  43. 43 Grok 4 60.5%
  44. 44 Qwen 3.8 27B 60.2%
  45. 45 Claude Opus 4.1 60%
  46. 46 Claude Haiku 5.5 59.2%
  47. 47 Claude Opus 4 58.8%
  48. 48 GLM-5.2 58.8%
  49. 49 Kimi K2.7 Code 57.9%
  50. 50 GPT-5.2 Pro xhigh 57.4%
  51. 51 GPT-5 high 56.7%
  52. 52 Grok 4.1 Fast 56%
  53. 53 GLM-5.1 55.1%
  54. 54 Claude Sonnet 4.5 54.3%
  55. 55 GPT-5.1 high 53.2%
  56. 56 GLM-5 53.2%
  57. 57 o3 high 53.1%
  58. 58 DeepSeek-V3.2-Speciale 52.6%
  59. 59 Gemini 2.5 Pro Mar 2025 51.6%
  60. 60 DeepSeek-V4-Pro 50.9%
  61. 61 Inkling 50%
  62. 62 GPT-5.6 Terra xhigh 48.9%
  63. 63 GLM-4.7 47.7%
  64. 64 Kimi K2.5 46.8%
  65. 65 GPT-5.6 Luna xhigh 46.8%
  66. 66 Claude 3.7 Sonnet thinking 46.4%
  67. 67 GPT-5.2 high 45.8%
  68. 68 MiniMax-M3 45.8%
  69. 69 Claude Sonnet 4 thinking 45.5%
  70. 70 o1-preview 41.7%
  71. 71 Nemotron 3 Ultra 41.7%
  72. 72 Claude 3.5 Sonnet (October 2024) 41.4%
  73. 73 Gemini 2.5 Flash Sep 2025 41.2%
  74. 74 DeepSeek-R1 May 2025 40.8%
  75. 75 o1 high 40.1%
  76. 76 DeepSeek-V3.1 40%
  77. 77 Kimi K2 Thinking 39.6%
  78. 78 o4-mini high 38.7%
  79. 79 Grok 3 36.1%
  80. 80 Qwen 3.6 Flash 35.2%
  81. 81 MiniMax-M2.1 34.7%
  82. 82 GPT-4.5 34.5%
  83. 83 Qwen3-235B-A22B 31%
  84. 84 DeepSeek-R1 30.9%
  85. 85 Gemini 2.0 Flash Thinking Jan 2025 30.7%
  86. 86 Llama 4 Maverick 27.7%
  87. 87 Claude 3.5 Sonnet 27.5%
  88. 88 DeepSeek-V3 Mar 2025 27.2%
  89. 89 Gemini 1.5 Pro Sep 2024 27.1%
  90. 90 GPT-4.1 27%
  91. 91 Kimi K2 Jul 2025 26.3%
  92. 92 GPT-4 Turbo Apr 2024 25.1%
  93. 93 MiniMax-M2 25%
  94. 94 Claude 3 Opus 23.5%
  95. 95 Llama 3.1-405B 23%
  96. 96 o3-mini high 22.8%
  97. 97 Grok-2 Aug 2024 22.7%
  98. 98 gpt-oss-120b 22.1%
  99. 99 Mistral Large 3 20.4%
  100. 100 Llama 3.3 70B Instruct 19.9%
  101. 101 Gemini 2.0 Flash Dec 2024 18.9%
  102. 102 DeepSeek-V3 18.9%
  103. 103 o1-mini 18.1%
  104. 104 GPT-4o Aug 2024 17.8%
  105. 105 Command R+ 17.4%
  106. 106 GPT-4o mini 10.7%
  107. 107 GPT-3.5 Turbo Jun 2023 8.00%

Sources: Official SimpleBench leaderboard and lmcouncil.ai, via Epoch AI (CC BY 4.0)

#EntryScoreSettingDateSource
1 Claude Fable 5.2 94.9% Nov 2026 Official leaderboard
2 Gemini 4 Carbon 92.6% Oct 2026 Official leaderboard
3 Gemini 4 Argon 91.7% Sep 2026 Official leaderboard
4 Claude Opus 5.5 88.4% Sep 2026 Official leaderboard
5 Claude Fable 5.1 86.6% Sep 2026 Official leaderboard
6 GPT-6 Astra Pro 86.5% Sep 2026 Official leaderboard
7 GPT-6 Astra 83.6% Sep 2026 Official leaderboard
8 GPT-6.1 Sol 82.9% Sep 2026 Official leaderboard
9 Gemini 3.8 Flash 82.4% Sep 2026 Official leaderboard
10 Claude Fable 5 81.9% Jun 2026 Official leaderboard
11 Muse Spark 1.3 81.8% Sep 2026 Official leaderboard
12 Claude Opus 5 80.6% Jul 2026 Official leaderboard
13 Gemini 3.1 Pro 79.6% Feb 2026 Official leaderboard
14 GPT-5.5 Pro 76.9% Apr 2026 Official leaderboard
15 Gemini 3.5 Flash 76.7% May 2026 Official leaderboard
16 Gemini 3 Pro 76.4% Nov 2025 Official leaderboard
17 Grok 4.6 75.9% Aug 2026 Official leaderboard
18 Claude Sonnet 5.5 75.9% Sep 2026 Official leaderboard
19 Muse Spark 1.2 74.5% Aug 2026 Official leaderboard
20 GPT-6 Sol 73.1% Sep 2026 Official leaderboard
21 Muse Ember 72.7% Nov 2026 Official leaderboard
22 GPT-6.1 Terra 72.2% Oct 2026 Official leaderboard
23 GPT-5.6 Sol Pro 71.7% xhigh Jul 2026 Official leaderboard
24 Qwen3.7-Max 70.4% May 2026 Official leaderboard
25 Grok 4.5 70% Jul 2026 Official leaderboard
26 GPT-5.5 69% Apr 2026 Official leaderboard
27 Claude Opus 4.6 67.6% Feb 2026 Official leaderboard
28 DeepSeek V4.1 Flash 66.7% Sep 2026 Official leaderboard
29 GLM-5.3 66.2% Aug 2026 Official leaderboard
30 Claude Opus 4.8 64.8% May 2026 Official leaderboard
31 GPT-5.6 Sol 64.8% xhigh Jul 2026 Official leaderboard
32 Qwen 3.6 Max (Preview) 63% Apr 2026 Official leaderboard
33 Qwen 3.8 Max 62.5% Aug 2026 Official leaderboard
34 Gemini 2.5 Pro Jun 2025 62.4% Jun 2025 Official leaderboard
35 Claude Opus 4.5 62% Nov 2025 Official leaderboard
36 Claude Opus 4.7 61.7% Apr 2026 Official leaderboard
37 GPT-5 Pro 61.6% Oct 2025 Official leaderboard
38 Gemini 3 Flash 61.1% Dec 2025 Official leaderboard
39 DeepSeek-V4-Flash 61.1% Apr 2026 Official leaderboard
40 Kimi K3 60.7% max Jul 2026 Official leaderboard
41 Claude Sonnet 5 60.6% Jun 2026 Official leaderboard
42 Grok 4 60.5% Jul 2025 Official leaderboard
43 Qwen 3.8 27B 60.2% Aug 2026 Official leaderboard
44 Claude Opus 4.1 60% Aug 2025 Official leaderboard
45 Claude Haiku 5.5 59.2% Oct 2026 Official leaderboard
46 Claude Opus 4 58.8% May 2025 Official leaderboard
47 GLM-5.2 58.8% Jun 2026 Official leaderboard
48 Kimi K2.7 Code 57.9% Jun 2026 Official leaderboard
49 GPT-5.2 Pro 57.4% xhigh Dec 2025 Official leaderboard
50 GPT-5 56.7% high Aug 2025 Official leaderboard
51 Grok 4.1 Fast 56% Nov 2025 Official leaderboard
52 GLM-5.1 55.1% Apr 2026 Official leaderboard
53 Ox Alpha 55.1% Aug 2026 Official leaderboard
54 Claude Sonnet 4.5 54.3% Sep 2025 Official leaderboard
55 GPT-5.1 53.2% high Nov 2025 Official leaderboard
56 GLM-5 53.2% Feb 2026 Official leaderboard
57 o3 53.1% high Apr 2025 Official leaderboard
58 DeepSeek-V3.2-Speciale 52.6% Dec 2025 Official leaderboard
59 Gemini 2.5 Pro Mar 2025 51.6% Mar 2025 Official leaderboard
60 DeepSeek-V4-Pro 50.9% Apr 2026 Official leaderboard
61 Inkling 50% Jul 2026 Official leaderboard
62 GPT-5.6 Terra 48.9% xhigh Jul 2026 Official leaderboard
63 GLM-4.7 47.7% Dec 2025 Official leaderboard
64 Kimi K2.5 46.8% Jan 2026 Official leaderboard
65 GPT-5.6 Luna 46.8% xhigh Jul 2026 Official leaderboard
66 Claude 3.7 Sonnet 46.4% thinking Feb 2025 Official leaderboard
67 DeepSeek V4 Flash (preview) 46.3% Jun 2026 Official leaderboard
68 GPT-5.2 45.8% high Dec 2025 Official leaderboard
69 MiniMax-M3 45.8% Jun 2026 Official leaderboard
70 Claude Sonnet 4 45.5% thinking May 2025 Official leaderboard
71 o1-preview 41.7% Sep 2024 Official leaderboard
72 Nemotron 3 Ultra 41.7% Jun 2026 Official leaderboard
73 Claude 3.5 Sonnet (October 2024) 41.4% Oct 2024 Official leaderboard
74 Gemini 2.5 Flash Sep 2025 41.2% Sep 2025 Official leaderboard
75 DeepSeek-R1 May 2025 40.8% May 2025 Official leaderboard
76 o1 40.1% high Dec 2024 Official leaderboard
77 DeepSeek-V3.1 40% Aug 2025 Official leaderboard
78 Kimi K2 Thinking 39.6% Nov 2025 Official leaderboard
79 o4-mini 38.7% high Apr 2025 Official leaderboard
80 Grok 3 36.1% Apr 2025 Official leaderboard
81 Qwen 3.6 Flash 35.2% Apr 2026 Official leaderboard
82 MiniMax-M2.1 34.7% Dec 2025 Official leaderboard
83 GPT-4.5 34.5% Feb 2025 Official leaderboard
84 Gemini-exp-1206 31.1% Dec 2024 Official leaderboard
85 Qwen3-235B-A22B 31% Apr 2025 Official leaderboard
86 DeepSeek-R1 30.9% Jan 2025 Official leaderboard
87 Gemini 2.0 Flash Thinking Jan 2025 30.7% Jan 2025 Official leaderboard
88 Llama 4 Maverick 27.7% Apr 2025 Official leaderboard
89 Claude 3.5 Sonnet 27.5% Jun 2024 Official leaderboard
90 DeepSeek-V3 Mar 2025 27.2% Mar 2025 Official leaderboard
91 Gemini 1.5 Pro Sep 2024 27.1% Sep 2024 Official leaderboard
92 GPT-4.1 27% Apr 2025 Official leaderboard
93 Kimi K2 Jul 2025 26.3% Jul 2025 Official leaderboard
94 GPT-4 Turbo Apr 2024 25.1% Apr 2024 Official leaderboard
95 MiniMax-M2 25% Oct 2025 Official leaderboard
96 Claude 3 Opus 23.5% Feb 2024 Official leaderboard
97 Llama 3.1-405B 23% Jul 2024 Official leaderboard
98 o3-mini 22.8% high Jan 2025 Official leaderboard
99 Grok-2 Aug 2024 22.7% Aug 2024 Official leaderboard
100 Mistral Large v2 22.5% Nov 2024 Official leaderboard
101 gpt-oss-120b 22.1% Aug 2025 Official leaderboard
102 Mistral Large 3 20.4% Dec 2025 Official leaderboard
103 Llama 3.3 70B Instruct 19.9% Dec 2024 Official leaderboard
104 Gemini 2.0 Flash Dec 2024 18.9% Dec 2024 Official leaderboard
105 DeepSeek-V3 18.9% Dec 2024 Official leaderboard
106 o1-mini 18.1% Sep 2024 Official leaderboard
107 GPT-4o Aug 2024 17.8% Aug 2024 Official leaderboard
108 Command R+ 17.4% Aug 2024 Official leaderboard
109 GPT-4o mini 10.7% Jul 2024 Official leaderboard
110 GPT-3.5 Turbo Jun 2023 8.00% Jun 2023 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Fable 5.2 94.9% Nov 2026 Official leaderboard
2 Gemini 4 Carbon 92.6% Oct 2026 Official leaderboard
3 Gemini 4 Argon 91.7% Sep 2026 Official leaderboard
4 Claude Opus 5.5 88.4% Sep 2026 Official leaderboard
5 Claude Fable 5.1 86.6% Sep 2026 Official leaderboard
6 GPT-6 Astra Pro 86.5% Sep 2026 Official leaderboard
7 GPT-6 Astra 83.6% Sep 2026 Official leaderboard
8 GPT-6.1 Sol 82.9% Sep 2026 Official leaderboard
9 Gemini 3.8 Flash 82.4% Sep 2026 Official leaderboard
10 Claude Fable 5 81.9% Jun 2026 Official leaderboard
11 Muse Spark 1.3 81.8% Sep 2026 Official leaderboard
12 Claude Opus 5 80.6% Jul 2026 Official leaderboard
13 Gemini 3.1 Pro 79.6% Feb 2026 Official leaderboard
14 GPT-5.5 Pro 76.9% Apr 2026 Official leaderboard
15 Gemini 3.5 Flash 76.7% May 2026 Official leaderboard
16 Gemini 3 Pro 76.4% Nov 2025 Official leaderboard
17 Grok 4.6 75.9% Aug 2026 Official leaderboard
18 Claude Sonnet 5.5 75.9% Sep 2026 Official leaderboard
19 Muse Spark 1.2 74.5% Aug 2026 Official leaderboard
20 GPT-6 Sol 73.1% Sep 2026 Official leaderboard
21 Muse Ember 72.7% Nov 2026 Official leaderboard
22 GPT-6.1 Terra 72.2% Oct 2026 Official leaderboard
23 GPT-5.6 Sol Pro 71.7% xhigh Jul 2026 Official leaderboard
24 GPT-5.6 Sol 71.7% pro, xhigh Jul 2026 Collected by Epoch AI, source not named
25 Qwen3.7-Max 70.4% May 2026 Official leaderboard
26 Grok 4.5 70% Jul 2026 Official leaderboard
27 GPT-5.5 69% Apr 2026 Official leaderboard
28 Claude Opus 4.6 67.6% Feb 2026 Official leaderboard
29 DeepSeek V4.1 Flash 66.7% Sep 2026 Official leaderboard
30 GLM-5.3 66.2% Aug 2026 Official leaderboard
31 Claude Opus 4.8 64.8% May 2026 Official leaderboard
32 Qwen 3.6 Max (Preview) 63% Apr 2026 Official leaderboard
33 Qwen 3.8 Max 62.5% Aug 2026 Official leaderboard
34 Gemini 2.5 Pro Jun 2025 62.4% Jun 2025 Official leaderboard
35 Claude Opus 4.5 62% Nov 2025 Official leaderboard
36 Claude Opus 4.7 61.7% Apr 2026 Official leaderboard
37 GPT-5 Pro 61.6% Oct 2025 Official leaderboard
38 Gemini 3 Flash 61.1% Dec 2025 Official leaderboard
39 DeepSeek-V4-Flash 61.1% Apr 2026 Official leaderboard
40 DeepSeek V4 Flash 0731 61.1% Jul 2026 Collected by Epoch AI, source not named
41 Kimi K3 60.7% max Jul 2026 Official leaderboard
42 Claude Sonnet 5 60.6% Jun 2026 Official leaderboard
43 Grok 4 60.5% Jul 2025 Official leaderboard
44 Qwen 3.8 27B 60.2% Aug 2026 Official leaderboard
45 Claude Opus 4.1 60% Aug 2025 Official leaderboard
46 Claude Haiku 5.5 59.2% Oct 2026 Official leaderboard
47 Claude Opus 4 58.8% May 2025 Official leaderboard
48 GLM-5.2 58.8% Jun 2026 Official leaderboard
49 Kimi K2.7 Code 57.9% Jun 2026 Official leaderboard
50 GPT-5.2 Pro 57.4% xhigh Dec 2025 Official leaderboard
51 GPT-5 56.7% high Aug 2025 Official leaderboard
52 Grok 4.1 Fast 56% Nov 2025 Official leaderboard
53 GLM-5.1 55.1% Apr 2026 Official leaderboard
54 Claude Sonnet 4.5 54.3% Sep 2025 Official leaderboard
55 GPT-5.1 53.2% high Nov 2025 Official leaderboard
56 GLM-5 53.2% Feb 2026 Official leaderboard
57 o3 53.1% high Apr 2025 Official leaderboard
58 DeepSeek-V3.2-Speciale 52.6% Dec 2025 Official leaderboard
59 Gemini 2.5 Pro Mar 2025 51.6% Mar 2025 Official leaderboard
60 DeepSeek-V4-Pro 50.9% Apr 2026 Official leaderboard
61 Inkling 50% Jul 2026 Official leaderboard
62 GPT-5.6 Terra 48.9% xhigh Jul 2026 Official leaderboard
63 GLM-4.7 47.7% Dec 2025 Official leaderboard
64 Kimi K2.5 46.8% Jan 2026 Official leaderboard
65 GPT-5.6 Luna 46.8% xhigh Jul 2026 Official leaderboard
66 Claude 3.7 Sonnet 46.4% thinking Feb 2025 Official leaderboard
67 GPT-5.2 45.8% high Dec 2025 Official leaderboard
68 MiniMax-M3 45.8% Jun 2026 Official leaderboard
69 Claude Sonnet 4 45.5% thinking May 2025 Official leaderboard
70 o1-preview 41.7% Sep 2024 Official leaderboard
71 Nemotron 3 Ultra 41.7% Jun 2026 Official leaderboard
72 Claude 3.5 Sonnet (October 2024) 41.4% Oct 2024 Official leaderboard
73 Gemini 2.5 Flash Sep 2025 41.2% Sep 2025 Official leaderboard
74 DeepSeek-R1 May 2025 40.8% May 2025 Official leaderboard
75 o1 40.1% high Dec 2024 Official leaderboard
76 DeepSeek-V3.1 40% Aug 2025 Official leaderboard
77 Kimi K2 Thinking 39.6% Nov 2025 Official leaderboard
78 o4-mini 38.7% high Apr 2025 Official leaderboard
79 Grok 3 36.1% Apr 2025 Official leaderboard
80 Qwen 3.6 Flash 35.2% Apr 2026 Official leaderboard
81 MiniMax-M2.1 34.7% Dec 2025 Official leaderboard
82 GPT-4.5 34.5% Feb 2025 Official leaderboard
83 Qwen3-235B-A22B 31% Apr 2025 Official leaderboard
84 DeepSeek-R1 30.9% Jan 2025 Official leaderboard
85 Gemini 2.0 Flash Thinking Jan 2025 30.7% Jan 2025 Official leaderboard
86 Llama 4 Maverick 27.7% Apr 2025 Official leaderboard
87 Claude 3.5 Sonnet 27.5% Jun 2024 Official leaderboard
88 DeepSeek-V3 Mar 2025 27.2% Mar 2025 Official leaderboard
89 Gemini 1.5 Pro Sep 2024 27.1% Sep 2024 Official leaderboard
90 GPT-4.1 27% Apr 2025 Official leaderboard
91 Kimi K2 Jul 2025 26.3% Jul 2025 Official leaderboard
92 GPT-4 Turbo Apr 2024 25.1% Apr 2024 Official leaderboard
93 MiniMax-M2 25% Oct 2025 Official leaderboard
94 Claude 3 Opus 23.5% Feb 2024 Official leaderboard
95 Llama 3.1-405B 23% Jul 2024 Official leaderboard
96 o3-mini 22.8% high Jan 2025 Official leaderboard
97 Grok-2 Aug 2024 22.7% Aug 2024 Official leaderboard
98 gpt-oss-120b 22.1% Aug 2025 Official leaderboard
99 Mistral Large 3 20.4% Dec 2025 Official leaderboard
100 Llama 3.3 70B Instruct 19.9% Dec 2024 Official leaderboard
101 Gemini 2.0 Flash Dec 2024 18.9% Dec 2024 Official leaderboard
102 DeepSeek-V3 18.9% Dec 2024 Official leaderboard
103 o1-mini 18.1% Sep 2024 Official leaderboard
104 GPT-4o Aug 2024 17.8% Aug 2024 Official leaderboard
105 Command R+ 17.4% Aug 2024 Official leaderboard
106 GPT-4o mini 10.7% Jul 2024 Official leaderboard
107 GPT-3.5 Turbo Jun 2023 8.00% Jun 2023 Official leaderboard

Epoch AI copied scores for 9 models from lmcouncil.ai. Epoch AI collected claims for 18 models from labs’ reports and papers. Switch the scores above to rank them together.

14 models have held number one on the official leaderboard since its release in October 2024 14 models have held number one with claims included since its release in October 2024

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 lmcouncil.ai, which Epoch AI copied.

OpenAI Anthropic Google DeepMind
Own leaderboard Passed 85% in Sep 2026, by Claude Fable 5.1
Claude Fable 5.2 94.9%
lmcouncil.ai, via Epoch AI Not tested since Feb 2026
Gemini 3.1 Pro 79.6%

Sources: Official SimpleBench leaderboard and lmcouncil.ai, via Epoch AI (CC BY 4.0)

Claude Fable 5.1 on the official leaderboard passed 85% in September 2026

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 On another leaderboard, collected by Epoch AI A claim Epoch AI collected from a report or paper Best so far
0% 20% 40% 60% 80% 100% 2023 2024 2025 2026 2027 Near the ceiling (85%) Random guessing Claude Fable 5.2 Claude Opus 5.5 Claude Fable 5 Gemini 3 Pro Gemini 2.5 Pro o3 Claude 3.7 Sonnet 0% 20% 40% 60% 80% 100% 2023 2024 2025 2026 2027 Near the ceiling (85%) Random guessing Claude Fable 5.2 Claude Opus 5.5

Sources: Official SimpleBench leaderboard, other boards, via Epoch AI (CC BY 4.0) and claims collected by Epoch AI (CC BY 4.0)

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 Science and reasoning

All 92 in Science and reasoning
  • GPQA Diamond In use 198 graduate-level 'Google-proof' multiple-choice questions in biology, chemistry and physics. Gemini 4 Carbon 96.5%
  • Terminal-Bench-Science In use 70 scientific research-workflow tasks across five science areas, run in Harbor containers. Hidden tests give the resolution rate. GPT-6 Astra 68.1%
  • FrontierScience In use Models solve physics, chemistry, and biology problems in an Olympiad track graded by short answer and a Research track graded by 10-point rubrics. Intern-S2-397B 87%
  • BioMysteryBench In use Bioinformatics problems on anonymized real datasets that require real analysis with tools. Rubric-graded, with human-solvable and human-difficult splits. Claude Fable 5.2 83.7%
  • HiPhO In use Models solve theory problems from 2024 and 2025 physics olympiad exams, some with diagrams. Official marking schemes grade answers and steps, mapped to medal levels. Gemini 3 Pro 78.3%
  • LABBench2 In use 11 biology research sub-tasks (literature, figures, tables, patents, trials and more) with tool access. Macro-averaged accuracy. GPT-5.2 60.9%