AMA

Value ranking

Best value on Terminal-Bench 2

Long-horizon shell-and-filesystem tasks executed in a sandboxed terminal, scored by whether the agent's final state matches a target state. Tests practical tool-using ability for everyday devops and data-wrangling work; one of the hardest agentic benchmarks today.

“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
    232.73
    69.8 / $0.30/M
  2. 2
    GPT-5 nano
    OpenAI
    139.00
    7.0 / $0.05/M
  3. 3
    DeepSeek V3
    DeepSeek
    123.26
    33.3 / $0.27/M
  4. 4
    GPT-5 mini
    OpenAI
    104.72
    26.2 / $0.25/M
  5. 5
    Gemini 3 Pro
    Google
    74.67
    93.3 / $1.25/M
  6. 6
    GPT-5.5
    OpenAI
    66.67
    100.0 / $1.50/M
  7. 7
    Kimi K2
    Moonshot (Kimi)
    64.35
    38.6 / $0.60/M
  8. 8
    GPT-5.4
    OpenAI
    63.81
    95.7 / $1.50/M
  9. 9
    GPT-5.2
    OpenAI
    56.57
    70.7 / $1.25/M
  10. 10
    GLM-4.7
    Zhipu AI (GLM)
    48.22
    24.1 / $0.50/M
  11. 11
    GPT-5
    OpenAI
    38.46
    48.1 / $1.25/M
  12. 12
    GPT-5.1
    OpenAI
    36.10
    45.1 / $1.25/M
  13. 13
    Claude Haiku 4.5
    Anthropic
    27.22
    27.2 / $1.00/M
  14. 14
    GLM-4.6
    Zhipu AI (GLM)
    21.90
    10.9 / $0.50/M
  15. 15
    Gemini 2.5 Pro
    Google
    18.34
    22.9 / $1.25/M
  16. 16
    Claude Sonnet 4.6
    Anthropic
    17.90
    53.7 / $3.00/M
  17. 17
    Claude Sonnet 4.5
    Anthropic
    12.67
    38.0 / $3.00/M
  18. 18
    Claude Opus 4.7
    Anthropic
    6.22
    93.3 / $15.00/M
  19. 19
    Claude Opus 4.6
    Anthropic
    6.18
    92.8 / $15.00/M
  20. 20
    Claude Opus 4.5
    Anthropic
    4.54
    68.0 / $15.00/M
  21. 21
    Grok 4
    xAI
    2.99
    14.9 / $5.00/M
  22. 22
    Claude Opus 4
    Anthropic
    2.06
    30.9 / $15.00/M
  23. 23
    Gemini 2.5 Flash
    Google
    0.00
    0.0 / $0.30/M
Open full leaderboard

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