Skip to content

Value ranking

Best value on SWE-bench Verified

As of October 25, 2026, the best value on SWE-bench Verified is GPT-5 nano, at 1468.00 normalized points per dollar of input price ($0.03 per million tokens, score 36.7).

Real GitHub issues solved end-to-end. Verified subset is a 500-task human-validated slice of SWE-bench.

“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-5 nano
    OpenAI
    1468.00
    36.7 / $0.03/M
  2. 2
    Qwen3 Coder 30B A3B Instruct
    Alibaba (Qwen)
    1045.71
    73.2 / $0.07/M
  3. 3
    GPT-5 mini
    OpenAI
    578.72
    72.3 / $0.13/M
  4. 4
    Gemini 2.0 Flash
    Google
    501.00
    50.1 / $0.10/M
  5. 5
    Gemini 3 Flash
    Google
    380.60
    95.2 / $0.25/M
  6. 6
    MiniMax M2.5
    MiniMax
    317.17
    95.2 / $0.30/M
  7. 7
    DeepSeek V3
    DeepSeek
    299.59
    86.9 / $0.29/M
  8. 8
    MiniMax M2
    MiniMax
    290.39
    74.0 / $0.26/M
  9. 9
    Qwen3 Coder 480B A35B
    Alibaba (Qwen)
    287.70
    86.3 / $0.30/M
  10. 10
    DeepSeek V3 (Thinking)
    DeepSeek
    250.45
    72.6 / $0.29/M
  11. 11
    Gemini 2.5 Flash
    Google
    186.93
    28.0 / $0.15/M
  12. 12
    gpt-oss-120b
    OpenAI
    161.00
    24.1 / $0.15/M
  13. 13
    Devstral 2 2512
    Mistral
    159.47
    63.8 / $0.40/M
  14. 14
    Kimi K2
    Moonshot (Kimi)
    155.42
    88.6 / $0.57/M
  15. 15
    GLM-4.6
    Zhipu AI (GLM)
    140.53
    84.3 / $0.60/M
  16. 16
    GLM 4.5
    Zhipu AI (GLM)
    131.02
    78.6 / $0.60/M
  17. 17
    GPT-5.1
    OpenAI
    129.89
    81.2 / $0.63/M
  18. 18
    GPT-5.2
    OpenAI
    103.86
    90.9 / $0.88/M
  19. 19
    Gemini 2.5 Pro
    Google
    101.60
    63.5 / $0.63/M
  20. 20
    Gemini 3.1 Pro
    Google
    97.43
    97.4 / $1.00/M
  21. 21
    Llama 4 Maverick
    Meta
    91.09
    17.1 / $0.19/M
  22. 22
    GLM 5
    Zhipu AI (GLM)
    90.88
    90.9 / $1.00/M
  23. 23
    Claude Haiku 4.5
    Anthropic
    82.04
    82.0 / $1.00/M
  24. 24
    GPT-5
    OpenAI
    74.53
    93.2 / $1.25/M
  25. 25
    o4-mini
    OpenAI
    71.98
    79.2 / $1.10/M
Open full leaderboard

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