AMA

Value ranking

Best value on ARC-AGI 2

Second-generation ARC challenge testing fluid reasoning over abstract visual puzzles. Resists training-data memorisation by construction: each puzzle is novel and solutions require multi-step pattern induction. Frontier models are only just starting to score above chance on the harder tier.

“Value” is normalized benchmark score (0–100 for this leaderboard cohort) divided by input price per million tokens. Higher means more capability per dollar on this axis only — always sanity-check latency, context length, and your real workload.

  1. 1
    Gemini 3 Flash
    Google
    131.80
    39.5 / $0.30/M
  2. 2
    Gemini 3 Pro
    Google
    72.57
    90.7 / $1.25/M
  3. 3
    GPT-5.5
    OpenAI
    66.67
    100.0 / $1.50/M
  4. 4
    GPT-5 nano
    OpenAI
    61.40
    3.1 / $0.05/M
  5. 5
    GPT-5.4
    OpenAI
    58.00
    87.0 / $1.50/M
  6. 6
    GPT-5.2
    OpenAI
    49.80
    62.3 / $1.25/M
  7. 7
    Claude Sonnet 4.6
    Anthropic
    23.69
    71.1 / $3.00/M
  8. 8
    Kimi K2
    Moonshot (Kimi)
    23.15
    13.9 / $0.60/M
  9. 9
    GPT-5 mini
    OpenAI
    20.88
    5.2 / $0.25/M
  10. 10
    DeepSeek V3
    DeepSeek
    17.56
    4.7 / $0.27/M
  11. 11
    GPT-5.1
    OpenAI
    16.60
    20.8 / $1.25/M
  12. 12
    Gemini 2.0 Flash
    Google
    15.30
    1.5 / $0.10/M
  13. 13
    GPT-5
    OpenAI
    9.28
    11.6 / $1.25/M
  14. 14
    o4-mini
    OpenAI
    6.54
    7.2 / $1.10/M
  15. 15
    Claude Opus 4.7
    Anthropic
    5.95
    89.2 / $15.00/M
  16. 16
    Claude Opus 4.6
    Anthropic
    5.43
    81.4 / $15.00/M
  17. 17
    Claude Sonnet 4.5
    Anthropic
    5.34
    16.0 / $3.00/M
  18. 18
    Claude Haiku 4.5
    Anthropic
    4.74
    4.7 / $1.00/M
  19. 19
    Grok 4
    xAI
    3.76
    18.8 / $5.00/M
  20. 20
    o3-mini
    OpenAI
    3.20
    3.5 / $1.10/M
  21. 21
    Claude Opus 4.5
    Anthropic
    2.95
    44.3 / $15.00/M
  22. 22
    DeepSeek R1
    DeepSeek
    2.78
    1.5 / $0.55/M
  23. 23
    o3
    OpenAI
    0.77
    7.7 / $10.00/M
  24. 24
    Gemini 1.5 Pro
    Google
    0.75
    0.9 / $1.25/M
  25. 25
    Claude Opus 4
    Anthropic
    0.68
    10.1 / $15.00/M
Open full leaderboard

AI Model Analyzer does not recommend specific vendors; rankings are derived from public data only.