Best value on SWE-bench Verified

“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
   648.00

   32.4 / $0.05/M
2. 2

   DeepSeek V3

   DeepSeek
   318.48

   86.0 / $0.27/M
3. 3

   Gemini 3 Flash

   Google
   316.07

   94.8 / $0.30/M
4. 4

   GPT-5 mini

   OpenAI
   281.84

   70.5 / $0.25/M
5. 5

   Kimi K2

   Moonshot (Kimi)
   146.37

   87.8 / $0.60/M
6. 6

   GLM-4.6

   Zhipu AI (GLM)
   127.52

   63.8 / $0.50/M
7. 7

   Claude Haiku 4.5

   Anthropic
   80.82

   80.8 / $1.00/M
8. 8

   Gemini 3 Pro

   Google
   77.81

   97.3 / $1.25/M
9. 9

   Gemini 2.5 Flash

   Google
   77.20

   23.2 / $0.30/M
10. 10

    GPT-5

    OpenAI
    74.15

    92.7 / $1.25/M
11. 11

    GPT-5.2

    OpenAI
    72.21

    90.3 / $1.25/M
12. 12

    o4-mini

    OpenAI
    70.70

    77.8 / $1.10/M
13. 13

    Gemini 2.5 Pro

    Google
    48.82

    61.0 / $1.25/M
14. 14

    o3-mini

    OpenAI
    39.97

    44.0 / $1.10/M
15. 15

    Claude Sonnet 4

    Anthropic
    32.12

    96.3 / $3.00/M
16. 16

    Claude Sonnet 4.5

    Anthropic
    31.10

    93.3 / $3.00/M
17. 17

    Claude 3.7 Sonnet

    Anthropic
    26.84

    80.5 / $3.00/M
18. 18

    Claude 3.5 Sonnet

    Anthropic
    25.31

    75.9 / $3.00/M
19. 19

    GPT-4o

    OpenAI
    15.40

    38.5 / $2.50/M
20. 20

    Claude Opus 4.5

    Anthropic
    6.67

    100.0 / $15.00/M
21. 21

    Claude Opus 4.6

    Anthropic
    6.30

    94.5 / $15.00/M
22. 22

    Claude Opus 4

    Anthropic
    6.06

    90.9 / $15.00/M
23. 23

    Claude 3 Opus

    Anthropic
    0.23

    3.5 / $15.00/M
24. 24

    Gemini 2.0 Flash

    Google
    0.00

    0.0 / $0.10/M

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