Skip to content
Leaders Board.
Boards/Reasoning

Humanity's Last Exam

2,500 expert-written questions across dozens of subjects, built to be the last closed-ended academic exam.
Scale AI and CAIS
GPT-6 Astra
#1 right now
54.8%
Accuracy
42
Models ranked
Sep 24, 2026
Board published

Ranking

Best variant per model, as published. Higher is better.

  1. 1
    GPT-6 Astra
    as “GPT 6 Astra”
    54.8%
  2. 2
    Claude Fable 5.1
    as “Fable 5.1 (xhigh)”
    46.5%
  3. 3
    Gemini 3.1 Pro Preview
    as “gemini-3.1-pro-preview (thinking high)”
    46.4%
  4. 4
    Gemini 3.8 Flash
    44.5%
  5. 5
    GPT-5.4 Pro
    as “gpt-5.4-pro-2026-03-05”
    44.3%
  6. 6
    Muse Spark
    40.6%
  7. 7
    Gemini 3 Pro
    as “gemini-3-pro-preview”
    37.5%
  8. 8
    GPT-5.4
    as “gpt-5.4-2026-03-05 (xhigh thinking)”
    36.2%
  9. 9
    Claude Opus 4.7
    as “claude-opus-4-7”
    36.2%
  10. 10
    Claude Opus 4.6
    as “claude-opus-4-6-thinking-max”
    34.4%
  11. 11
    GPT-5 Pro
    as “gpt-5-pro-2025-10-06”
    31.6%
  12. 12
    GPT-5.2
    as “gpt-5.2-2025-12-11”
    27.8%
  13. 13
    GPT-5
    as “gpt-5-2025-08-07”
    25.3%
  14. 14
    Claude Opus 4.5
    as “claude-opus-4-5-20251101-thinking”
    25.2%
  15. 15
    Kimi K2.5
    as “kimi-k2.5”
    24.4%
  16. 16
    GPT-5.1
    as “gpt-5.1-thinking”
    23.7%
  17. 17
    Gemini 2.5 Pro (Jun 2025)
    as “gemini-2.5-pro-preview-06-05”
    21.6%
  18. 18
    o3
    as “o3 (high) (April 2025)”
    20.3%
  19. 19
    GPT-5 mini
    as “gpt-5-mini-2025-08-07”
    19.4%
  20. 20
    o4-mini
    as “o4-mini (high) (April 2025)”
    18.1%
  21. 21
    Claude Sonnet 4.5
    as “claude-sonnet-4-5-20250929-thinking”
    13.7%
  22. 22
    Gemini 2.5 Flash (Sep 2025)
    as “Gemini 2.5 Flash (April 2025)”
    12.1%
  23. 23
    Claude Opus 4.1
    as “claude-opus-4-1-20250805-thinking”
    11.5%
  24. 24
    Claude Opus 4
    as “Claude Opus 4 (Thinking)”
    10.7%
  25. 25
    Gemini 3.1 Flash-Lite
    as “gemini-3.1-flash-lite-preview”
    8.6%

Rank over time

History builds with every daily capture. One capture so far.

Other reasoning boards

AA Intelligence
#1 Claude Opus 5.5
GPQA Diamond
#1 GPT-6 Astra
Epoch ECI
#1 GPT-6 Astra
ARC-AGI-2
#1 GPT-6 Astra
LiveBench
#1 Claude Fable 5.1
MMLU-Pro
#1 Gemini 3.1 Pro Preview
Open LLM LB
#1 none

Scores as published by Scale AI and CAIS on the capture date (2026-09-24). Source: scale.com leaderboard page.

scale.com/leaderboard/humanitys_last_exam

Weekly: the AI leaderboards with a new #1, Saturday mornings.