FrontierMath

Epoch AI leaderboard on unpublished, highly challenging mathematics problems, from undergraduate level (Tiers 1–3) to research level (Tier 4).

At a glance

Claude Fable 5.2 (Max, Default Fallback) leads with 95.1%, out of 128 results. The problems are unpublished and their answers checked automatically, so a model cannot have seen them in training. Each row is one reasoning setting, and two scores whose 95% intervals overlap do not separate.

Data retrieved on Nov 22, 2026.

View on Epoch AI
Tier

From undergraduate level to exploratory problems for an advanced graduate student. The home page's Math ranking follows this tier.

Runs
128
Best score
95.1%
Claude Fable 5.2 (Max, Default Fallback)
Newest
Gemini 4 Neon (Xhigh)
Nov 18, 2026 · 91.2%
Version
v2
Released on 12 June 2026

128 / 128 runs

  1. #1
    Claude Fable 5.2max

    Anthropic

    95.1%

    Released: Nov 9, 2026Std. error: ± 1.3%

  2. #2
    GPT-6.1 Solmax

    OpenAI

    93.7%

    Released: Sep 29, 2026Std. error: ± 1.4%

  3. #3
    GPT-6 Astramax

    OpenAI

    93.7%

    Released: Sep 3, 2026Std. error: ± 1.4%

  4. #4
    GPT-6.2 Astramax

    OpenAI

    92.6%

    Released: Nov 17, 2026Std. error: ± 1.6%

  5. #5
    Gemini 4 Neonxhigh

    Google

    91.2%

    Released: Nov 18, 2026Std. error: ± 1.7%

  6. #6
    Claude Opus 5.5max

    Anthropic

    91.2%

    Released: Sep 22, 2026Std. error: ± 1.7%

  7. #7
    GPT-6.1 Terramax

    OpenAI

    90.9%

    Released: Oct 26, 2026Std. error: ± 1.7%

  8. #8
    Gemini 4 Carbonhigh

    Google

    90.5%

    Released: Nov 10, 2026Std. error: ± 1.7%

  9. #9
    Claude Fable 5.1max

    Anthropic

    90.2%

    Released: Sep 1, 2026Std. error: ± 1.8%

  10. #10
    GPT-6 Solmax

    OpenAI

    89.8%

    Released: Sep 22, 2026Std. error: ± 1.8%

  11. #11
    GPT-5.6 Solmax

    OpenAI

    89.1%

    Released: Jul 9, 2026Std. error: ± 1.8%

  12. #12
    Claude Sonnet 5.5max

    Anthropic

    88.8%

    Released: Sep 28, 2026Std. error: ± 1.9%

  13. #1388.1%

    Released: Nov 12, 2026Std. error: ± 1.9%

  14. #14
    GPT-5.5 Proxhigh

    OpenAI

    87.7%

    Released: Apr 23, 2026Std. error: ± 1.9%

  15. #15
    Claude Fable 5max

    Anthropic

    87.0%

    Released: Jun 9, 2026Std. error: ± 2.0%

  16. #16
    Gemini 4 Argonhigh

    Google

    86.0%

    Released: Oct 13, 2026Std. error: ± 2.1%

  17. #17
    GPT-5.6 Terramax

    OpenAI

    86.0%

    Released: Jul 9, 2026Std. error: ± 2.1%

  18. #18
    Claude Opus 5max

    Anthropic

    85.6%

    Released: Jul 24, 2026Std. error: ± 2.1%

  19. #19
    GPT-5.5xhigh

    OpenAI

    85.3%

    Released: Apr 23, 2026Std. error: ± 2.1%

  20. #20
    GPT-5.4 Proxhigh

    OpenAI

    82.5%

    Released: Mar 5, 2026Std. error: ± 2.3%

  21. #2182.1%

    Released: Oct 12, 2026Std. error: ± 2.3%

  22. #22
    GPT-5.6 Lunamax

    OpenAI

    82.1%

    Released: Jul 9, 2026Std. error: ± 2.3%

  23. #23
    Claude Opus 4.8max

    Anthropic

    80.0%

    Released: May 28, 2026Std. error: ± 2.4%

  24. #24
    GPT-6 Lunamax

    OpenAI

    78.9%

    Released: Sep 22, 2026Std. error: ± 2.4%

  25. #25
    GPT-5.4xhigh

    OpenAI

    78.6%

    Released: Mar 5, 2026Std. error: ± 2.4%

  26. #26
    Kimi K3.1 Promax

    Moonshot AI

    78.2%

    Released: Oct 23, 2026Std. error: ± 2.4%

  27. #27
    Claude Haiku 5.5max

    Anthropic

    75.1%

    Released: Oct 7, 2026Std. error: ± 2.6%

  28. #28
    Qwen 3.8 Maxxhigh

    Alibaba

    74.7%

    Released: Aug 2, 2026Std. error: ± 2.6%

  29. #2974.4%

    Released: Sep 2, 2026Std. error: ± 2.6%

  30. #3074.0%

    Released: Sep 2, 2026Std. error: ± 2.6%

  31. #31
    GPT-5.2 Proxhigh

    OpenAI

    74.0%

    Released: Dec 11, 2025Std. error: ± 2.6%

  32. #32
    Kimi K3max

    Moonshot AI

    72.2%

    Released: Jul 16, 2026Std. error: ± 2.7%

  33. #3371.6%

    Released: Aug 13, 2026Std. error: ± 2.7%

  34. #3471.2%

    Released: Oct 17, 2026Std. error: ± 2.7%

  35. #35
    Claude Opus 4.7max

    Anthropic

    70.2%

    Released: Apr 16, 2026Std. error: ± 2.7%

  36. #3669.8%

    Released: Oct 15, 2026Std. error: ± 2.7%

  37. #37
    GLM-5.3max

    Z.AI

    68.8%

    Released: Aug 14, 2026Std. error: ± 2.7%

  38. #3868.4%

    Released: Sep 2, 2026Std. error: ± 2.8%

  39. #39
    GPT-5.2xhigh

    OpenAI

    67.4%

    Released: Dec 11, 2025Std. error: ± 2.8%

  40. #4067.4%

    Released: Sep 9, 2026Std. error: ± 2.8%

  41. #41
    Grok 4.6xhigh

    SpaceXSI

    66.0%

    Released: Aug 12, 2026Std. error: ± 2.8%

  42. #42
    Claude Opus 4.6max

    Anthropic

    66.0%

    Released: Feb 5, 2026Std. error: ± 2.8%

  43. #43
    Qwen3.8 Max (0902)xhigh

    Alibaba

    65.6%

    Released: Sep 1, 2026Std. error: ± 2.8%

  44. #44
    Claude Sonnet 5max

    Anthropic

    65.6%

    Released: Jun 30, 2026Std. error: ± 2.8%

  45. #45
    DeepSeek V4 Pro 0813max

    DeepSeek

    64.6%

    Released: Aug 13, 2026Std. error: ± 2.8%

  46. #4664.6%

    Released: May 19, 2026Std. error: ± 2.8%

  47. #4762.8%

    Released: May 19, 2026Std. error: ± 2.9%

  48. #4859.6%

    Released: Feb 19, 2026Std. error: ± 2.9%

  49. #49
    GLM-5.2max

    Z.AI

    59.2%

    Released: Jun 16, 2026Std. error: ± 3.0%

  50. #5058.9%

    Released: Jul 21, 2026Std. error: ± 2.9%

  51. #5158.6%

    Released: Oct 6, 2026Std. error: ± 2.9%

  52. #52
    DeepSeek V4 Flash 0731max

    DeepSeek

    57.5%

    Released: Jul 31, 2026Std. error: ± 2.9%

  53. #53
    Grok 4.5high

    SpaceXSI

    57.2%

    Released: Jul 8, 2026Std. error: ± 2.9%

  54. #54
    Kimi K2.6

    Moonshot

    57.2%

    Released: Apr 20, 2026Std. error: ± 2.9%

  55. #5555.8%

    Released: Aug 20, 2026Std. error: ± 2.9%

  56. #56
    GPT-5 Prohigh

    OpenAI

    55.8%

    Released: Oct 7, 2025Std. error: ± 2.9%

  57. #57
    GPT-5high

    OpenAI

    55.4%

    Released: Aug 7, 2025Std. error: ± 2.9%

  58. #58
    GLM-5.2low

    Z.ai (Zhipu AI)

    54.7%

    Released: Jun 16, 2026Std. error: ± 3.0%

  59. #59
    Kimi K2.7 Code

    Moonshot AI

    54.0%

    Released: Jun 12, 2026Std. error: ± 3.0%

  60. #60
    Grok 4.7xhigh

    SpaceXSI

    53.0%

    Released: Sep 21, 2026Std. error: ± 3.0%

  61. #61
    GPT-5.4 Minixhigh

    OpenAI

    51.2%

    Released: Mar 17, 2026Std. error: ± 3.0%

  62. #6251.2%

    Released: Dec 17, 2025Std. error: ± 3.0%

  63. #63
    GPT-5 minihigh

    OpenAI

    46.7%

    Released: Aug 7, 2025Std. error: ± 3.0%

  64. #64
    Inkling-Smallxhigh

    Thinking Machines

    46.3%

    Released: Jul 15, 2026Std. error: ± 3.0%

  65. #65
    DeepSeek-V4-Promax

    DeepSeek

    45.3%

    Released: Apr 24, 2026Std. error: ± 3.0%

  66. #66
    Grok 4.20

    xAI

    44.9%

    Released: Feb 17, 2026Std. error: ± 3.0%

  67. #67
    GPT-5.4 Nanohigh

    OpenAI

    44.9%

    Released: Mar 17, 2026Std. error: ± 3.0%

  68. #68
    Grok 4.3 Betahigh

    SpaceXSI

    42.8%

    Released: Apr 17, 2026Std. error: ± 2.9%

  69. #69
    GLM-5.2

    Z.ai (Zhipu AI)

    42.5%

    Released: Jun 16, 2026Std. error: ± 2.9%

  70. #70
    GPT-5.6 Lunalow

    OpenAI

    41.4%

    Released: Jul 9, 2026Std. error: ± 2.9%

  71. #71
    GPT-5.6 Luna

    OpenAI

    39.6%

    Released: Jul 9, 2026Std. error: ± 2.9%

  72. #7238.2%

    Released: Mar 31, 2026Std. error: ± 2.9%

  73. #73
    GPT-5low

    OpenAI

    37.2%

    Released: Aug 7, 2025Std. error: ± 2.9%

  74. #74
    GLM-5.1

    Z.ai (Zhipu AI)

    36.8%

    Released: Apr 7, 2026Std. error: ± 2.9%

  75. #75
    o4-minihigh

    OpenAI

    36.1%

    Released: Apr 16, 2025Std. error: ± 2.9%

  76. #76
    Qwen3.6 27B

    Alibaba

    35.1%

    Released: Apr 22, 2026Std. error: ± 2.8%

  77. #7734.4%

    Released: Jun 2, 2026Std. error: ± 2.8%

  78. #78
    Claude Opus 4.5

    Anthropic

    34.4%

    Released: Nov 24, 2025Std. error: ± 2.8%

  79. #79
    Qwen3.6 27B

    Alibaba

    34.0%

    Released: Apr 22, 2026Std. error: ± 2.8%

  80. #80
    o3high

    OpenAI

    33.3%

    Released: Apr 16, 2025Std. error: ± 2.8%

  81. #81
    Inklingxhigh

    Thinking Machines

    33.3%

    Released: Jul 15, 2026Std. error: ± 2.8%

  82. #8232.3%

    Released: Mar 31, 2026Std. error: ± 2.8%

  83. #83
    Qwen3.5 397B-A17B

    Alibaba

    31.2%

    Released: Feb 13, 2026Std. error: ± 2.7%

  84. #84
    o3medium

    OpenAI

    29.8%

    Released: Apr 16, 2025Std. error: ± 2.7%

  85. #85
    Qwen3.5 397B-A17B

    Alibaba

    29.5%

    Released: Feb 13, 2026Std. error: ± 2.7%

  86. #86
    o4-minimedium

    OpenAI

    28.8%

    Released: Apr 16, 2025Std. error: ± 2.7%

  87. #8727.7%

    Released: Mar 3, 2026Std. error: ± 2.7%

  88. #88
    GPT-5.5 Instant

    OpenAI

    26.3%

    Released: May 5, 2026Std. error: ± 2.6%

  89. #8926.0%

    Released: Jul 21, 2026Std. error: ± 2.6%

  90. #90
    GLM-5.1

    Z.ai (Zhipu AI)

    24.9%

    Released: Apr 7, 2026Std. error: ± 2.6%

  91. #91
    GPT-5.4 Minilow

    OpenAI

    24.6%

    Released: Mar 17, 2026Std. error: ± 2.6%

  92. Released: Jun 17, 2025Std. error: ± 2.6%

  93. #93
    Claude Sonnet 4.5

    Anthropic

    23.9%

    Released: Sep 29, 2025Std. error: ± 2.5%

  94. #9422.5%

    Released: Mar 3, 2026Std. error: ± 2.5%

  95. #95
    Qwen 3.6 Flash

    Alibaba

    22.5%

    Released: Apr 27, 2026Std. error: ± 2.5%

  96. #9621.4%

    Released: Mar 3, 2026Std. error: ± 2.4%

  97. #97
    Qwen 3.6 35B-A3B

    Alibaba

    20.4%

    Released: Apr 14, 2026Std. error: ± 2.4%

  98. #98
    GPT-5.4 Nanolow

    OpenAI

    20.4%

    Released: Mar 17, 2026Std. error: ± 2.4%

  99. #99
    GPT-5 nanohigh

    OpenAI

    20.0%

    Released: Aug 7, 2025Std. error: ± 2.4%

  100. #100
    Qwen3.7 Flash

    Alibaba

    19.3%

    Released: Jul 27, 2026Std. error: ± 2.3%

  101. #101
    Qwen3.7 Flash

    Alibaba

    19.3%

    Released: Jul 27, 2026Std. error: ± 2.3%

  102. #102
    o3low

    OpenAI

    19.3%

    Released: Apr 16, 2025Std. error: ± 2.3%

  103. #103
    Qwen3-Max

    Alibaba

    18.9%

    Released: Sep 24, 2025Std. error: ± 2.3%

  104. #104
    o3-minihigh

    OpenAI

    18.6%

    Released: Jan 31, 2025Std. error: ± 2.3%

  105. #105
    Qwen 3.5 Flash (hosted 35B-A3B)

    Alibaba

    18.2%

    Released: Feb 25, 2026Std. error: ± 2.3%

  106. #106
    GPT-5minimal

    OpenAI

    18.2%

    Released: Aug 7, 2025Std. error: ± 2.3%

  107. #107
    GPT-5 minilow

    OpenAI

    18.2%

    Released: Aug 7, 2025Std. error: ± 2.3%

  108. #108
    Qwen 3.6 35B-A3B

    Alibaba

    17.5%

    Released: Apr 14, 2026Std. error: ± 2.3%

  109. #109
    Qwen 3.6 Flash

    Alibaba

    17.2%

    Released: Apr 27, 2026Std. error: ± 2.2%

  110. #110
    GPT-5.4 Mini

    OpenAI

    17.2%

    Released: Mar 17, 2026Std. error: ± 2.2%

  111. #111
    o4-minilow

    OpenAI

    16.1%

    Released: Apr 16, 2025Std. error: ± 2.2%

  112. #112
    o1high

    OpenAI

    14.7%

    Released: Dec 17, 2024Std. error: ± 2.1%

  113. #113
    Claude Opus 4.1

    Anthropic

    12.6%

    Released: Aug 5, 2025Std. error: ± 2.0%

  114. #114
    o3-minimedium

    OpenAI

    10.5%

    Released: Jan 31, 2025Std. error: ± 1.8%

  115. #115
    o1medium

    OpenAI

    10.2%

    Released: Dec 17, 2024Std. error: ± 1.8%

  116. #116
    Qwen 3.5 Flash (hosted 35B-A3B)

    Alibaba

    9.5%

    Released: Feb 25, 2026Std. error: ± 1.7%

  117. #117
    o1low

    OpenAI

    8.4%

    Released: Dec 17, 2024Std. error: ± 1.6%

  118. #1186.7%

    Released: Apr 14, 2025Std. error: ± 1.5%

  119. #119
    GPT-4.1

    OpenAI

    6.0%

    Released: Apr 14, 2025Std. error: ± 1.4%

  120. #120
    GPT-5 miniminimal

    OpenAI

    6.0%

    Released: Aug 7, 2025Std. error: ± 1.4%

  121. #121
    GPT-5 nanolow

    OpenAI

    6.0%

    Released: Aug 7, 2025Std. error: ± 1.4%

  122. #122
    GPT-5.4 Nano

    OpenAI

    4.6%

    Released: Mar 17, 2026Std. error: ± 1.2%

  123. #123
    o3-minilow

    OpenAI

    3.9%

    Released: Jan 31, 2025Std. error: ± 1.1%

  124. #124
    GPT-5 nanominimal

    OpenAI

    1.8%

    Released: Aug 7, 2025Std. error: ± 0.8%

  125. #1250.7%

    Released: Jul 18, 2024Std. error: ± 0.5%

  126. Released: Apr 9, 2024Std. error: ± 0.5%

  127. #1270.4%

    Released: Aug 6, 2024Std. error: ± 0.4%

  128. #128
    GPT-3.5 Turbo (Jan 2024)

    OpenAI

    0.0%

    Released: Jan 25, 2024Std. error: ± 0.0%

Method and versions

FrontierMath is an Epoch AI benchmark of several hundred unpublished mathematics problems. Tiers 1 to 3 range from undergraduate problems to exploratory problems suited to an advanced graduate student; Tier 4 is research-level mathematics. BenchSift shows version 2, released on 12 June 2026, which fixed errors in 42% of problems.

  • Scores are on the private v2 sets; earlier versions are not mixed into these results.
  • Epoch AI evaluates each model at one or more reasoning settings (low to max), each on its own row.
  • The standard error is Epoch AI's; the score bar shows the 95% interval it implies.
  • A run is tied to a catalogue page by its model id and setting; a run without a matching page keeps Epoch AI's name.
  • Epoch AI states that FrontierMath is supported by OpenAI.
  • Epoch AI's data is published under the Creative Commons Attribution 4.0 licence.

Data via Epoch AI, FrontierMath v2 (CC BY 4.0).

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy