Skip to content

Value ranking

Best value on Recovery Rate

As of October 25, 2026, the best value on Recovery Rate is GPT-6 Luna, at 1800.40 normalized points per dollar of input price ($0.05 per million tokens, score 90.0).

How often the model self-corrects after producing an incorrect intermediate step (debugging axis upstream). Critical for agentic loops that depend on the model noticing and repairing its own mistakes rather than barrelling forward.

“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
    GPT-6 Luna
    OpenAI
    1800.40
    90.0 / $0.05/M
  2. 2
    DeepSeek V4 Flash 0423
    DeepSeek
    948.24
    86.3 / $0.09/M
  3. 3
    GPT-5.6 Luna
    OpenAI
    794.60
    79.5 / $0.10/M
  4. 4
    Gemini 3.5 Flash Lite
    Google
    642.87
    96.4 / $0.15/M
  5. 5
    DeepSeek V4 Pro 0423
    DeepSeek
    457.18
    95.5 / $0.21/M
  6. 6
    Gemini 3.8 Flash
    Google
    266.67
    100.0 / $0.38/M
  7. 7
    GPT-6 Sol
    OpenAI
    100.00
    100.0 / $1.00/M
  8. 8
    GPT-5.6 Terra
    OpenAI
    100.00
    100.0 / $1.00/M
  9. 9
    Kimi K2.7 Code
    Moonshot (Kimi)
    93.62
    88.9 / $0.95/M
  10. 10
    GLM 5.3
    Zhipu AI (GLM)
    68.25
    95.5 / $1.40/M
  11. 11
    GPT-5.6 Sol
    OpenAI
    50.00
    100.0 / $2.00/M
  12. 12
    Claude Sonnet 5
    Anthropic
    43.92
    87.8 / $2.00/M
  13. 13
    Claude Sonnet 5.5
    Anthropic
    41.49
    83.0 / $2.00/M
  14. 14
    Kimi K3
    Moonshot (Kimi)
    33.33
    100.0 / $3.00/M
  15. 15
    GPT-6.1 Sol
    OpenAI
    28.22
    28.2 / $1.00/M
  16. 16
    Claude Opus 5.5
    Anthropic
    21.48
    85.9 / $4.00/M
  17. 17
    GPT-6 Astra
    OpenAI
    20.00
    100.0 / $5.00/M
  18. 18
    Claude Opus 5
    Anthropic
    20.00
    100.0 / $5.00/M
  19. 19
    GPT-5.5
    OpenAI
    11.54
    28.9 / $2.50/M
  20. 20
    Claude Sonnet 4.6
    Anthropic
    10.64
    31.9 / $3.00/M
  21. 21
    Claude Fable 5.1
    Anthropic
    10.00
    100.0 / $10.00/M
  22. 22
    Claude Opus 4.7
    Anthropic
    4.93
    24.7 / $5.00/M
  23. 23
    Claude Opus 4.8
    Anthropic
    0.33
    1.6 / $5.00/M
  24. 24
    Claude Opus 4.6
    Anthropic
    0.00
    0.0 / $5.00/M
Open full leaderboard

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