Benchmark Almanac

Conceptual Reasoning Index

Claude Fable 5.2 79.0 leads since November 2026
Not in model cards

An index that ranks models on reasoning about hard-to-verify topics such as AI alignment and decision theory. It weights LMCA 60 percent and ACCoRD and DTBench 20 percent each.

The leaderboard CSV had 267 model rows in November 2026.

Released
August 2026
Kind
Composite indices
Level
Expert
Scored
Checked automatically
Answers
Short answer
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
154
Entries on its own leaderboard, 151 models we track

Claude Fable 5.2 leads with 79.0

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.

154 entries. Scroll sideways for the rest.

  1. 79.0 Claude Fable 5.2
  2. 77.9 Claude Opus 5.5
  3. 77.1 Claude Sonnet 5.5
  4. 74.7 Claude Fable 5.1
  5. 73.8 GPT-6 Astra
  6. 73.6 Claude Opus 5
  7. 72.7 GPT-6.1 Sol
  8. 72.7 Claude Fable 5
  9. 72.3 Gemini 4 Argon
  10. 71.0 GPT-6 Sol
  11. 70.0 GPT-5.6 Sol Pro
  12. 68.7 GPT-5.6 Sol
  13. 68.3 Claude Opus 4.8
  14. 67.8 Kimi K3.1 Pro
  15. 67.7 Muse Wonder
  16. 67.3 GPT-5.5
  17. 66.6 Claude Opus 4.6
  18. 66.2 GPT-5.5 Pro
  19. 66.2 Muse Spark 1.3
  20. 65.7 GPT-5.6 Terra
  21. 65.5 Claude Opus 4.7
  22. 65.3 GPT-5.4
  23. 65.3 Claude Haiku 5.5
  24. 65.0 Gemini 3.8 Flash
  25. 64.6 Gemini 3.7 Flash
  26. 63.8 Gemini 3.1 Pro
  27. 63.8 Muse Spark 1.1
  28. 63.4 GLM-5.3
  29. 63.1 Grok 4.7
  30. 62.8 Kimi K3
  31. 62.7 Muse Spark 1.2
  32. 62.5 Grok 4.6
  33. 61.8 Claude Sonnet 5
  34. 60.3 Grok 4.5
  35. 59.8 Gemini 3.5 Flash
  36. 59.5 Gemini 3.6 Flash
  37. 59.4 Qwen 3.8 Max
  38. 59.4 Qwen3.8-2.4T-A95B
  39. 59.3 DeepSeek V4 Pro 0813
  40. 59.2 Claude Sonnet 4.6
  41. 59.2 GPT-5.2
  42. 59.1 Claude Opus 4.5
  43. 58.7 GPT-5.6 Luna
  44. 58.6 GLM-5.2
  45. 58.4 DeepSeek V4.1 Flash
  46. 57.8 GPT-6 Luna
  47. 57.3 GPT-5.1
  48. 56.9 Qwen3.7-Max
  49. 56.0 Gemini 3 Flash
  50. 55.8 DeepSeek-V4-Pro
  51. 55.6 Grok 4.20 Multi-agent
  52. 55.5 GPT-5
  53. 55.5 Qwen 3.8 27B
  54. 55.4 Qwen 3.6 Max (Preview)
  55. 54.4 DeepSeek V4 Flash 0731
  56. 53.7 Grok 4.20
  57. 53.6 Kimi K2.6
  58. 53.5 o3-pro
  59. 53.0 o3
  60. 52.9 Grok 4.3
  61. 52.6 Inkling
  62. 52.2 Tencent Hy3
  63. 51.9 GPT-5.4 Mini
  64. 51.8 Claude Sonnet 4.5
  65. 51.8 Nemotron 3 Ultra
  66. 51.0 GPT-5.2 Instant
  67. 50.7 Mistral Large 4
  68. 50.7 Gemini 3.5 Flash-Lite
  69. 50.5 Muse Glimmer 30B
  70. 50.2 Gemma 4 31B IT
  71. 49.7 Qwen3.5 397B-A17B
  72. 49.5 Claude Opus 4
  73. 49.4 Qwen3.7-Plus
  74. 49.0 GPT-5.4 Nano
  75. 48.6 Qwen3-Max
  76. 48.5 DeepSeek-V4-Flash
  77. 48.3 GPT-5 mini
  78. 48.3 Claude Opus 4.1
  79. 48.1 GPT-5.3 Instant
  80. 47.5 Gemini 2.5 Pro
  81. 46.6 Qwen3.5-27B
  82. 46.5 Qwen 3.5 Plus (hosted 397B-A17B)
  83. 46.4 MiniMax-M3
  84. 46.1 Gemini 3.1 Flash-Lite
  85. 45.7 DeepSeek-V3.2-Exp
  86. 45.7 Qwen3.5-122B-A10B
  87. 45.7 Qwen 3.6 Plus
  88. 45.4 DeepSeek-V3.2
  89. 45.2 Qwen3.6 27B
  90. 44.2 MiMo-V2.5-Pro
  91. 43.3 Qwen 3.6 Flash
  92. 43.1 Claude Haiku 4.5
  93. 42.4 Qwen3-235B-A22B-Thinking
  94. 42.3 Qwen 3.5 Flash (hosted 35B-A3B)
  95. 42.1 Qwen3.5-35B-A3B
  96. 42.0 DeepSeek-V3.1-Terminus
  97. 41.3 Qwen Plus
  98. 41.2 o4-mini
  99. 41.1 Claude Sonnet 4
  100. 40.3 Gemma 4 26B A4B
  101. 39.9 DeepSeek-V3.1
  102. 39.7 Qwen 3.6 35B-A3B
  103. 38.9 gpt-oss-120b
  104. 38.7 Qwen3 235B-A22B (2507)
  105. 37.8 o1
  106. 37.7 Gemini 2.5 Flash
  107. 36.7 Qwen3-Next-80B-A3B
  108. 36.6 Mistral Medium 3.5
  109. 36.3 Qwen3-235B-A22B
  110. 36.0 GPT-4.1
  111. 35.1 Qwen3.5-9B
  112. 33.6 Nemotron 3.5 Lightning 30B-A3B
  113. 32.9 Qwen3-30B-A3B-Thinking
  114. 32.3 o3-mini
  115. 31.6 Qwen3-30B-A3B-Instruct
  116. 31.0 GPT-4.1 mini
  117. 30.5 Mistral Small 4
  118. 30.4 gpt-oss-20b
  119. 29.2 Qwen3-32B
  120. 28.6 Qwen3-14B
  121. 28.4 Mistral Medium 3.1
  122. 28.1 Mistral Large 3
  123. 27.1 GPT-4o
  124. 26.8 Llama 4 Maverick
  125. 26.2 DeepSeek-V3
  126. 26.1 Llama 3.3 70B
  127. 25.3 GPT-4
  128. 25.2 Qwen3-30B-A3B
  129. 25.1 Gemini 2.5 Flash-Lite
  130. 25.0 DeepSeek Chat
  131. 24.2 Llama 3.1-70B
  132. 24.2 Qwen2.5-72B
  133. 24.1 GPT-5 nano
  134. 24.0 Claude 3 Opus
  135. 23.1 Llama 4 Scout
  136. 21.5 GPT-4 Turbo
  137. 20.7 Cohere Command A
  138. 20.2 Qwen3-8B
  139. 19.6 Gemma 3 27B
  140. 19.3 Ministral 3 14B
  141. 17.5 GPT-4o mini
  142. 17.4 Ministral 3 8B Instruct
  143. 13.6 Llama 3.1-8B
  144. 13.3 Gemma 3n E4B
  145. 13.3 Gemma 3 12B
  146. 12.7 GPT-3.5 Turbo
  147. 12.3 GPT-4.1 nano
  148. 11.9 Qwen2.5-7B
  149. 11.6 Claude 3 Haiku
  150. 11.6 Command R+
  151. 10.3 c4ai-command-r-08-2024
  152. 10.1 Ministral 3B
  153. 10.1 Gemma 3 4B
  154. 7.1 Command R7B
  1. 1 Claude Fable 5.2 max 79.0
  2. 2 Claude Opus 5.5 max 77.9
  3. 3 Claude Sonnet 5.5 max 77.1
  4. 4 Claude Fable 5.1 max 74.7
  5. 5 GPT-6 Astra max 73.8
  6. 6 Claude Opus 5 max 73.6
  7. 7 GPT-6.1 Sol 72.7
  8. 8 Claude Fable 5 max 72.7
  9. 9 Gemini 4 Argon high 72.3
  10. 10 GPT-6 Sol 71.0
  11. 11 GPT-5.6 Sol Pro 70.0
  12. 12 GPT-5.6 Sol max 68.7
  13. 13 Claude Opus 4.8 68.3
  14. 14 Kimi K3.1 Pro 67.8
  15. 15 Muse Wonder 67.7
  16. 16 GPT-5.5 67.3
  17. 17 Claude Opus 4.6 66.6
  18. 18 GPT-5.5 Pro 66.2
  19. 19 Muse Spark 1.3 66.2
  20. 20 GPT-5.6 Terra max 65.7
  21. 21 Claude Opus 4.7 65.5
  22. 22 GPT-5.4 65.3
  23. 23 Claude Haiku 5.5 max 65.3
  24. 24 Gemini 3.8 Flash high 65.0
  25. 25 Gemini 3.7 Flash high 64.6
  26. 26 Gemini 3.1 Pro high 63.8
  27. 27 Muse Spark 1.1 63.8
  28. 28 GLM-5.3 63.4
  29. 29 Grok 4.7 63.1
  30. 30 Kimi K3 62.8
  31. 31 Muse Spark 1.2 62.7
  32. 32 Grok 4.6 62.5
  33. 33 Claude Sonnet 5 max 61.8
  34. 34 Grok 4.5 60.3
  35. 35 Gemini 3.5 Flash 59.8
  36. 36 Gemini 3.6 Flash 59.5
  37. 37 Qwen 3.8 Max 59.4
  38. 38 Qwen3.8-2.4T-A95B 59.4
  39. 39 DeepSeek V4 Pro 0813 59.3
  40. 40 Claude Sonnet 4.6 59.2
  41. 41 GPT-5.2 59.2
  42. 42 Claude Opus 4.5 59.1
  43. 43 GPT-5.6 Luna max 58.7
  44. 44 GLM-5.2 58.6
  45. 45 DeepSeek V4.1 Flash 58.4
  46. 46 GPT-6 Luna 57.8
  47. 47 GPT-5.1 57.3
  48. 48 Qwen3.7-Max 56.9
  49. 49 Gemini 3 Flash 56.0
  50. 50 DeepSeek-V4-Pro 55.8
  51. 51 Grok 4.20 Multi-agent 55.6
  52. 52 GPT-5 55.5
  53. 53 Qwen 3.8 27B 55.5
  54. 54 Qwen 3.6 Max (Preview) 55.4
  55. 55 DeepSeek V4 Flash 0731 54.4
  56. 56 Grok 4.20 53.7
  57. 57 Kimi K2.6 53.6
  58. 58 o3-pro 53.5
  59. 59 o3 53.0
  60. 60 Grok 4.3 52.9
  61. 61 Inkling 52.6
  62. 62 Tencent Hy3 52.2
  63. 63 GPT-5.4 Mini 51.9
  64. 64 Claude Sonnet 4.5 51.8
  65. 65 Nemotron 3 Ultra 51.8
  66. 66 GPT-5.2 Instant 51.0
  67. 67 Mistral Large 4 50.7
  68. 68 Gemini 3.5 Flash-Lite 50.7
  69. 69 Muse Glimmer 30B 50.5
  70. 70 Gemma 4 31B IT 50.2
  71. 71 Qwen3.5 397B-A17B 49.7
  72. 72 Claude Opus 4 49.5
  73. 73 Qwen3.7-Plus 49.4
  74. 74 GPT-5.4 Nano 49.0
  75. 75 Qwen3-Max thinking 48.6
  76. 76 DeepSeek-V4-Flash 48.5
  77. 77 GPT-5 mini 48.3
  78. 78 Claude Opus 4.1 48.3
  79. 79 GPT-5.3 Instant 48.1
  80. 80 Gemini 2.5 Pro Jun 2025 47.5
  81. 81 Qwen3.5-27B 46.6
  82. 82 Qwen 3.5 Plus (hosted 397B-A17B) 46.5
  83. 83 MiniMax-M3 46.4
  84. 84 Gemini 3.1 Flash-Lite 46.1
  85. 85 DeepSeek-V3.2-Exp 45.7
  86. 86 Qwen3.5-122B-A10B 45.7
  87. 87 Qwen 3.6 Plus 45.7
  88. 88 DeepSeek-V3.2 45.4
  89. 89 Qwen3.6 27B 45.2
  90. 90 MiMo-V2.5-Pro 44.2
  91. 91 Qwen 3.6 Flash 43.3
  92. 92 Claude Haiku 4.5 43.1
  93. 93 Qwen3-235B-A22B-Thinking Jul 2025 42.4
  94. 94 Qwen 3.5 Flash (hosted 35B-A3B) 42.3
  95. 95 Qwen3.5-35B-A3B 42.1
  96. 96 DeepSeek-V3.1-Terminus 42.0
  97. 97 Qwen Plus Apr 2025 41.3
  98. 98 o4-mini 41.2
  99. 99 Claude Sonnet 4 41.1
  100. 100 Gemma 4 26B A4B 40.3
  101. 101 DeepSeek-V3.1 non-thinking 39.9
  102. 102 Qwen 3.6 35B-A3B 39.7
  103. 103 gpt-oss-120b 38.9
  104. 104 Qwen3 235B-A22B (2507) 38.7
  105. 105 o1 37.8
  106. 106 Gemini 2.5 Flash Jun 2025 37.7
  107. 107 Qwen3-Next-80B-A3B thinking 36.7
  108. 108 Mistral Medium 3.5 36.6
  109. 109 Qwen3-235B-A22B 36.3
  110. 110 GPT-4.1 36.0
  111. 111 Qwen3.5-9B 35.1
  112. 112 Nemotron 3.5 Lightning 30B-A3B 33.6
  113. 113 Qwen3-30B-A3B-Thinking Jul 2025 32.9
  114. 114 o3-mini 32.3
  115. 115 Qwen3-30B-A3B-Instruct Jul 2025 31.6
  116. 116 GPT-4.1 mini 31.0
  117. 117 Mistral Small 4 30.5
  118. 118 gpt-oss-20b 30.4
  119. 119 Qwen3-32B 29.2
  120. 120 Qwen3-14B 28.6
  121. 121 Mistral Medium 3.1 28.4
  122. 122 Mistral Large 3 28.1
  123. 123 GPT-4o May 2024 27.1
  124. 124 Llama 4 Maverick 26.8
  125. 125 DeepSeek-V3 Mar 2025 26.2
  126. 126 Llama 3.3 70B 26.1
  127. 127 GPT-4 Jun 2023 25.3
  128. 128 Qwen3-30B-A3B 25.2
  129. 129 Gemini 2.5 Flash-Lite Jun 2025 25.1
  130. 130 DeepSeek Chat 25.0
  131. 131 Llama 3.1-70B 24.2
  132. 132 Qwen2.5-72B 24.2
  133. 133 GPT-5 nano 24.1
  134. 134 Claude 3 Opus 24.0
  135. 135 Llama 4 Scout 23.1
  136. 136 GPT-4 Turbo Apr 2024 21.5
  137. 137 Cohere Command A 20.7
  138. 138 Qwen3-8B 20.2
  139. 139 Gemma 3 27B 19.6
  140. 140 Ministral 3 14B 19.3
  141. 141 GPT-4o mini 17.5
  142. 142 Ministral 3 8B Instruct 17.4
  143. 143 Llama 3.1-8B 13.6
  144. 144 Gemma 3n E4B 13.3
  145. 145 Gemma 3 12B 13.3
  146. 146 GPT-3.5 Turbo Nov 2023 12.7
  147. 147 GPT-4.1 nano 12.3
  148. 148 Qwen2.5-7B 11.9
  149. 149 Claude 3 Haiku 11.6
  150. 150 Command R+ 11.6
  151. 151 c4ai-command-r-08-2024 10.3
  152. 152 Ministral 3B 10.1
  153. 153 Gemma 3 4B 10.1
  154. 154 Command R7B 7.1

Source: Official Conceptual Reasoning Index leaderboard

#EntryScoreSettingDateSource
1 Claude Fable 5.2 79.0 max Nov 2026 Official leaderboard
2 Claude Opus 5.5 77.9 max Sep 2026 Official leaderboard
3 Claude Sonnet 5.5 77.1 max Sep 2026 Official leaderboard
4 Claude Fable 5.1 74.7 max Sep 2026 Official leaderboard
5 GPT-6 Astra 73.8 max Sep 2026 Official leaderboard
6 Claude Opus 5 73.6 max Jul 2026 Official leaderboard
7 GPT-6.1 Sol 72.7 Sep 2026 Official leaderboard
8 Claude Fable 5 72.7 max Jun 2026 Official leaderboard
9 Gemini 4 Argon 72.3 high Sep 2026 Official leaderboard
10 GPT-6 Sol 71.0 Sep 2026 Official leaderboard
11 GPT-5.6 Sol Pro 70.0 Jul 2026 Official leaderboard
12 GPT-5.6 Sol 68.7 max Jul 2026 Official leaderboard
13 Claude Opus 4.8 68.3 May 2026 Official leaderboard
14 Kimi K3.1 Pro 67.8 Oct 2026 Official leaderboard
15 Muse Wonder 67.7 Oct 2026 Official leaderboard
16 GPT-5.5 67.3 Apr 2026 Official leaderboard
17 Claude Opus 4.6 66.6 Feb 2026 Official leaderboard
18 GPT-5.5 Pro 66.2 Apr 2026 Official leaderboard
19 Muse Spark 1.3 66.2 Sep 2026 Official leaderboard
20 GPT-5.6 Terra 65.7 max Jul 2026 Official leaderboard
21 Claude Opus 4.7 65.5 Apr 2026 Official leaderboard
22 GPT-5.4 65.3 Mar 2026 Official leaderboard
23 Claude Haiku 5.5 65.3 max Oct 2026 Official leaderboard
24 Gemini 3.8 Flash 65.0 high Sep 2026 Official leaderboard
25 Gemini 3.7 Flash 64.6 high Aug 2026 Official leaderboard
26 Gemini 3.1 Pro 63.8 high Feb 2026 Official leaderboard
27 Muse Spark 1.1 63.8 Jul 2026 Official leaderboard
28 GLM-5.3 63.4 Aug 2026 Official leaderboard
29 Grok 4.7 63.1 Sep 2026 Official leaderboard
30 Kimi K3 62.8 Jul 2026 Official leaderboard
31 Muse Spark 1.2 62.7 Aug 2026 Official leaderboard
32 Grok 4.6 62.5 Aug 2026 Official leaderboard
33 Claude Sonnet 5 61.8 max Jun 2026 Official leaderboard
34 Grok 4.5 60.3 Jul 2026 Official leaderboard
35 Gemini 3.5 Flash 59.8 May 2026 Official leaderboard
36 Gemini 3.6 Flash 59.5 Jul 2026 Official leaderboard
37 Qwen 3.8 Max 59.4 Aug 2026 Official leaderboard
38 Qwen3.8-2.4T-A95B 59.4 Aug 2026 Official leaderboard
39 DeepSeek V4 Pro 0813 59.3 Aug 2026 Official leaderboard
40 Claude Sonnet 4.6 59.2 Feb 2026 Official leaderboard
41 GPT-5.2 59.2 Dec 2025 Official leaderboard
42 Claude Opus 4.5 59.1 Nov 2025 Official leaderboard
43 GPT-5.6 Luna 58.7 max Jul 2026 Official leaderboard
44 GLM-5.2 58.6 Jun 2026 Official leaderboard
45 DeepSeek V4.1 Flash 58.4 Sep 2026 Official leaderboard
46 GPT-6 Luna 57.8 Sep 2026 Official leaderboard
47 GPT-5.1 57.3 Nov 2025 Official leaderboard
48 Qwen3.7-Max 56.9 May 2026 Official leaderboard
49 Gemini 3 Flash 56.0 Dec 2025 Official leaderboard
50 DeepSeek-V4-Pro 55.8 Apr 2026 Official leaderboard
51 Grok 4.20 Multi-agent 55.6 Mar 2026 Official leaderboard
52 GPT-5 55.5 Aug 2025 Official leaderboard
53 Qwen 3.8 27B 55.5 Aug 2026 Official leaderboard
54 Qwen 3.6 Max (Preview) 55.4 Apr 2026 Official leaderboard
55 DeepSeek V4 Flash 0731 54.4 Jul 2026 Official leaderboard
56 Grok 4.20 53.7 Feb 2026 Official leaderboard
57 Kimi K2.6 53.6 Apr 2026 Official leaderboard
58 o3-pro 53.5 Jun 2025 Official leaderboard
59 o3 53.0 Apr 2025 Official leaderboard
60 Grok 4.3 52.9 Apr 2026 Official leaderboard
61 Inkling 52.6 Jul 2026 Official leaderboard
62 Tencent Hy3 52.2 Jul 2026 Official leaderboard
63 GPT-5.4 Mini 51.9 Mar 2026 Official leaderboard
64 Claude Sonnet 4.5 51.8 Sep 2025 Official leaderboard
65 Nemotron 3 Ultra 51.8 Jun 2026 Official leaderboard
66 GPT-5.2 Instant 51.0 Dec 2025 Official leaderboard
67 Mistral Large 4 50.7 Oct 2026 Official leaderboard
68 Gemini 3.5 Flash-Lite 50.7 Jul 2026 Official leaderboard
69 Muse Glimmer 30B 50.5 Aug 2026 Official leaderboard
70 Gemma 4 31B IT 50.2 Apr 2026 Official leaderboard
71 Qwen3.5 397B-A17B 49.7 Feb 2026 Official leaderboard
72 Claude Opus 4 49.5 May 2025 Official leaderboard
73 Qwen3.7-Plus 49.4 Jun 2026 Official leaderboard
74 GPT-5.4 Nano 49.0 Mar 2026 Official leaderboard
75 Qwen3-Max 48.6 thinking Sep 2025 Official leaderboard
76 DeepSeek-V4-Flash 48.5 Apr 2026 Official leaderboard
77 GPT-5 mini 48.3 Aug 2025 Official leaderboard
78 Claude Opus 4.1 48.3 Aug 2025 Official leaderboard
79 GPT-5.3 Instant 48.1 Mar 2026 Official leaderboard
80 Gemini 2.5 Pro Jun 2025 47.5 Jun 2025 Official leaderboard
81 Qwen3.5-27B 46.6 Feb 2026 Official leaderboard
82 Qwen 3.5 Plus (hosted 397B-A17B) 46.5 Feb 2026 Official leaderboard
83 MiniMax-M3 46.4 Jun 2026 Official leaderboard
84 Gemini 3.1 Flash-Lite 46.1 Mar 2026 Official leaderboard
85 DeepSeek-V3.2-Exp 45.7 Sep 2025 Official leaderboard
86 Qwen3.5-122B-A10B 45.7 Feb 2026 Official leaderboard
87 Qwen 3.6 Plus 45.7 Mar 2026 Official leaderboard
88 DeepSeek-V3.2 45.4 Dec 2025 Official leaderboard
89 Qwen3.6 27B 45.2 Apr 2026 Official leaderboard
90 MiMo-V2.5-Pro 44.2 Apr 2026 Official leaderboard
91 Qwen 3.6 Flash 43.3 Apr 2026 Official leaderboard
92 Claude Haiku 4.5 43.1 Oct 2025 Official leaderboard
93 Qwen3-235B-A22B-Thinking Jul 2025 42.4 Jul 2025 Official leaderboard
94 Qwen 3.5 Flash (hosted 35B-A3B) 42.3 Feb 2026 Official leaderboard
95 Qwen3.5-35B-A3B 42.1 Feb 2026 Official leaderboard
96 DeepSeek-V3.1-Terminus 42.0 Sep 2025 Official leaderboard
97 Qwen Plus Apr 2025 41.3 Apr 2025 Official leaderboard
98 o4-mini 41.2 Apr 2025 Official leaderboard
99 Claude Sonnet 4 41.1 May 2025 Official leaderboard
100 Gemma 4 26B A4B 40.3 Apr 2026 Official leaderboard
101 DeepSeek-V3.1 39.9 non-thinking Aug 2025 Official leaderboard
102 Qwen 3.6 35B-A3B 39.7 Apr 2026 Official leaderboard
103 gpt-oss-120b 38.9 Aug 2025 Official leaderboard
104 Qwen3 235B-A22B (2507) 38.7 Jul 2025 Official leaderboard
105 o1 37.8 Dec 2024 Official leaderboard
106 Gemini 2.5 Flash Jun 2025 37.7 Jun 2025 Official leaderboard
107 Qwen3-Next-80B-A3B 36.7 thinking Sep 2025 Official leaderboard
108 Mistral Medium 3.5 36.6 Apr 2026 Official leaderboard
109 Qwen3-235B-A22B 36.3 Apr 2025 Official leaderboard
110 GPT-4.1 36.0 Apr 2025 Official leaderboard
111 Qwen3.5-9B 35.1 Feb 2026 Official leaderboard
112 Nemotron 3.5 Lightning 30B-A3B 33.6 Aug 2026 Official leaderboard
113 Qwen3-30B-A3B-Thinking Jul 2025 32.9 Jul 2025 Official leaderboard
114 o3-mini 32.3 Jan 2025 Official leaderboard
115 Qwen3-30B-A3B-Instruct Jul 2025 31.6 Jul 2025 Official leaderboard
116 GPT-4.1 mini 31.0 Apr 2025 Official leaderboard
117 Mistral Small 4 30.5 Mar 2026 Official leaderboard
118 gpt-oss-20b 30.4 Aug 2025 Official leaderboard
119 Qwen3-32B 29.2 Apr 2025 Official leaderboard
120 Qwen3-14B 28.6 Apr 2025 Official leaderboard
121 Mistral Medium 3.1 28.4 Aug 2025 Official leaderboard
122 Mistral Large 3 28.1 Dec 2025 Official leaderboard
123 GPT-4o May 2024 27.1 May 2024 Official leaderboard
124 Llama 4 Maverick 26.8 Apr 2025 Official leaderboard
125 DeepSeek-V3 Mar 2025 26.2 Mar 2025 Official leaderboard
126 Llama 3.3 70B 26.1 Dec 2024 Official leaderboard
127 GPT-4 Jun 2023 25.3 Jun 2023 Official leaderboard
128 Qwen3-30B-A3B 25.2 Apr 2025 Official leaderboard
129 Gemini 2.5 Flash-Lite Jun 2025 25.1 Jun 2025 Official leaderboard
130 DeepSeek Chat 25.0 Dec 2024 Official leaderboard
131 Llama 3.1-70B 24.2 Jul 2024 Official leaderboard
132 Qwen2.5-72B 24.2 Sep 2024 Official leaderboard
133 GPT-5 nano 24.1 Aug 2025 Official leaderboard
134 Claude 3 Opus 24.0 Feb 2024 Official leaderboard
135 Llama 4 Scout 23.1 Apr 2025 Official leaderboard
136 GPT-4 Turbo Apr 2024 21.5 Apr 2024 Official leaderboard
137 Cohere Command A 20.7 Mar 2025 Official leaderboard
138 Qwen3-8B 20.2 Apr 2025 Official leaderboard
139 Gemma 3 27B 19.6 Mar 2025 Official leaderboard
140 Ministral 3 14B 19.3 Dec 2025 Official leaderboard
141 GPT-4o mini 17.5 Jul 2024 Official leaderboard
142 Ministral 3 8B Instruct 17.4 Dec 2025 Official leaderboard
143 Llama 3.1-8B 13.6 Jul 2024 Official leaderboard
144 Gemma 3n E4B 13.3 Jun 2025 Official leaderboard
145 Gemma 3 12B 13.3 Mar 2025 Official leaderboard
146 GPT-3.5 Turbo Nov 2023 12.7 Nov 2023 Official leaderboard
147 GPT-4.1 nano 12.3 Apr 2025 Official leaderboard
148 Qwen2.5-7B 11.9 Sep 2024 Official leaderboard
149 Claude 3 Haiku 11.6 Mar 2024 Official leaderboard
150 Command R+ 11.6 Aug 2024 Official leaderboard
151 c4ai-command-r-08-2024 10.3 Aug 2024 Official leaderboard
152 Ministral 3B 10.1 Oct 2024 Official leaderboard
153 Gemma 3 4B 10.1 Mar 2025 Official leaderboard
154 Command R7B 7.1 Dec 2024 Official leaderboard

4 models have held number one on the official leaderboard since its release in August 2026

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

Anthropic
Own leaderboard 4 leaders
Claude Fable 5.2 79.0

Source: Official Conceptual Reasoning Index 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
0 20 40 60 80 2023 2024 2025 2026 2027 0 20 40 60 80 2023 2024 2025 2026 2027

Source: Official Conceptual Reasoning Index 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 Other

All 43 in Other
  • Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
  • Artificial Analysis Intelligence Index v5 In use Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness. Gemini 4 Carbon 59.0
  • SAGE v1.1 (Vals) In use Models grade handwritten undergraduate math and computer science work against rubrics. Scored by class-balanced accuracy of each rubric check against expert grading. Claude Opus 5.5 91.3%
  • MTEB Fading Massive Text Embedding Benchmark: retrieval, clustering, classification across languages and modalities. sonar 42%
  • Anthropic ECI (AECI) Only in system cards Index that combines a basket of capability benchmarks into one capability score to track trends across Anthropic models. Claude Opus 5.5 169.4
  • SPEED-Bench Only in system cards Measures speculative decoding across varied prompts and domains. Reports the average acceptance length of draft tokens, so higher means faster generation at the same output. Nemotron 3 Ultra 5.15