Humanity's Last Exam

Humanity's Last Exam

Source

Artificial Analysis

Version

Not reported

Measurement date

Not reported

Metric

% · Higher is better

Evaluation scope

Broad expert-level knowledge evaluation; observed 2026-10-19; result date not supplied

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

1. #1Claude Opus 5.5

   anthropic/claude-opus-5-5

   61.4%

   View modelEvidence independent

2. #2Claude Fable 5.1

   anthropic/claude-fable-5-1

   59.1%

   View modelEvidence independent

3. #3Gemini 4 Argon

   google/gemini-4-argon-20261013

   57.1%

   View modelEvidence independent

4. #4Claude Fable 5

   anthropic/claude-fable-5

   55.5%

   View modelEvidence independent

5. #5Claude Sonnet 5.5

   anthropic/claude-sonnet-5-5

   55%

   View modelEvidence independent

6. #6Claude Opus 5

   anthropic/claude-opus-5

   54.9%

   View modelEvidence independent

7. #7GPT-6 Astra

   openai/gpt-6-astra-20260903

   54.7%

   View modelEvidence independent

8. #8GPT-6.1 Sol

   openai/gpt-6.1-sol-20260929

   52.9%

   View modelEvidence independent

9. #9Muse Wonder

   meta/muse-wonder-20261012

   50.1%

   View modelEvidence independent

10. #10GPT-5.6 Sol

    openai/gpt-5.6-sol-20260709

    49.5%

    View modelEvidence independent

11. #11MiMo-V2.6-Pro

    xiaomi/mimo-v2.6-pro-20260921

    49.4%

    View modelEvidence independent

12. #12Muse Spark 1.3

    meta/muse-spark-1.3-20260902

    48.7%

    View modelEvidence independent

13. #13Gemini 3.7 Flash

    google/gemini-3.7-flash-20260813

    47.9%

    View modelEvidence independent

14. #13GPT-6 Sol

    openai/gpt-6-sol-20260922

    47.9%

    View modelEvidence independent

15. #15Gemini 3.8 Flash

    google/gemini-3.8-flash-20260902

    47.8%

    View modelEvidence independent

16. #16Kimi K3

    moonshotai/kimi-k3-20260715

    46.9%

    View modelEvidence independent

17. #17Step 5 Preview

    stepfun/step-5-preview-20261008

    46.5%

    View modelEvidence independent

18. #18Muse Spark 1.2

    meta/muse-spark-1.2-20260805

    45.5%

    View modelEvidence independent

19. #19Claude Haiku 5.5

    anthropic/claude-haiku-5-5

    44.4%

    View modelEvidence independent

20. #20Muse Spark Open

    meta/muse-spark-open-20261015

    43.9%

    View modelEvidence independent

21. #21DeepSeek V4.1 Flash 1017

    deepseek/deepseek-v4.1-flash-20261017

    43.3%

    View modelEvidence independent

22. #22Grok 4.7

    x-ai/grok-4.7-20260916

    43.1%

    View modelEvidence independent

23. #22Qwen3.8 Max (0902)

    qwen/qwen3.8-max-20260902

    43.1%

    View modelEvidence independent

24. #24GPT-5.6 Terra

    openai/gpt-5.6-terra-20260709

    42.9%

    View modelEvidence independent

25. #24Grok 4.6

    x-ai/grok-4.6-20260810

    42.9%

    View modelEvidence independent

26. #26GLM 5.3

    z-ai/glm-5.3-20260816

    42.3%

    View modelEvidence independent

27. #27DeepSeek V4 Pro 0813

    deepseek/deepseek-v4-pro-20260813

    41%

    View modelEvidence independent

28. #28GLM 5.3 Flash

    z-ai/glm-5.3-flash-20260826

    39.9%

    View modelEvidence independent

29. #29GPT-5.6 Luna

    openai/gpt-5.6-luna-20260709

    39.5%

    View modelEvidence independent

30. #30DeepSeek V4.1 Flash

    deepseek/deepseek-v4.1-flash-20260910

    39.2%

    View modelEvidence independent

31. #31MiniMax M3

    minimax/minimax-m3-20260531

    39%

    View modelEvidence independent

32. #32GPT-6 Luna

    openai/gpt-6-luna-20260922

    38.5%

    View modelEvidence independent

33. #33MiMo-V2.5-Pro

    xiaomi/mimo-v2.5-pro-20260422

    35.7%

    View modelEvidence independent

34. #34Mistral Large 4

    mistralai/mistral-large-4-0-20261006

    35%

    View modelEvidence independent

35. #35Qwen3.8 27B

    qwen/qwen3.8-27b-20260814

    33.9%

    View modelEvidence independent

36. #36Hy3

    tencent/hy3-20260706

    33.5%

    View modelEvidence independent

37. #36Hy3 preview

    tencent/hy3-preview-20260421

    33.5%

    View modelEvidence independent

38. #38Inkling

    thinkingmachines/inkling-20260715

    31.9%

    View modelEvidence independent

39. #39Nemotron 3.5 Super VL

    nvidia/nemotron-3.5-super-vl-20261015

    30.7%

    View modelEvidence independent

40. #40Nemotron 3 Ultra

    nvidia/nemotron-3-ultra-550b-a55b-20260604

    28.4%

    View modelEvidence independent

41. #41MiMo-V2.5

    xiaomi/mimo-v2.5-20260422

    27.2%

    View modelEvidence independent

42. #42gpt-oss-120b

    openai/gpt-oss-120b

    19.6%

    View modelEvidence independent

43. #43Gemini 3.5 Flash Lite

    google/gemini-3.5-flash-lite-20260721

    18.8%

    View modelEvidence independent

44. #44Mistral Medium 3

    mistralai/mistral-medium-3

    13.8%

    View modelEvidence independent

45. #44Mistral Medium 3.1

    mistralai/mistral-medium-3.1

    13.8%

    View modelEvidence independent

46. #44Mistral Medium 3.5

    mistralai/mistral-medium-3.5-20260430

    13.8%

    View modelEvidence independent

47. #47Command A

    cohere/command-a-03-2025

    12%

    View modelEvidence independent

48. #47Command A+

    cohere/command-a-plus-05-2026

    12%

    View modelEvidence independent

49. #49Nemotron 3.5 Lightning

    nvidia/nemotron-3.5-lightning-20260807

    10.6%

    View modelEvidence independent

50. #50MiMo-V2-Flash

    xiaomi/mimo-v2-flash

    8.6%

    View modelEvidence independent

51. #51Phi 4

    microsoft/phi-4

    3.8%

    View modelEvidence independent