Skip to content

Value ranking

Best value on MMMU-Pro

As of October 25, 2026, the best value on MMMU-Pro is Gemma 4 26B A4B, at 553.53 normalized points per dollar of input price ($0.15 per million tokens, score 83.0).

Harder MMMU variant: text-only-answerable questions removed, ten answer options instead of four, and a vision-only setting where the question is embedded in the image.

“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
    Gemma 4 26B A4B
    Google
    553.53
    83.0 / $0.15/M
  2. 2
    Gemma 4 31B
    Google
    450.70
    90.1 / $0.20/M
  3. 3
    Seed 1.6
    ByteDance Seed
    264.24
    66.1 / $0.25/M
  4. 4
    GPT-5.1
    OpenAI
    151.92
    95.0 / $0.63/M
  5. 5
    GPT-5.2
    OpenAI
    112.19
    98.2 / $0.88/M
  6. 6
    Gemini 2.5 Pro
    Google
    111.55
    69.7 / $0.63/M
  7. 7
    Gemini 3.1 Pro
    Google
    98.39
    98.4 / $1.00/M
  8. 8
    GPT-5.4
    OpenAI
    80.00
    100.0 / $1.25/M
  9. 9
    GPT-5
    OpenAI
    46.06
    57.6 / $1.25/M
  10. 10
    o3
    OpenAI
    44.49
    89.0 / $2.00/M
  11. 11
    Claude Sonnet 4.6
    Anthropic
    28.21
    84.6 / $3.00/M
  12. 12
    Qwen2.5 VL 72B Instruct
    Alibaba (Qwen)
    24.65
    19.7 / $0.80/M
  13. 13
    Claude Sonnet 4.5
    Anthropic
    23.93
    71.8 / $3.00/M
  14. 14
    Gemini 1.5 Pro
    Google
    17.06
    21.3 / $1.25/M
  15. 15
    Claude Opus 4.6
    Anthropic
    16.65
    83.3 / $5.00/M
  16. 16
    Claude Opus 4.5
    Anthropic
    16.65
    83.3 / $5.00/M
  17. 17
    GPT-4o
    OpenAI
    13.12
    32.8 / $2.50/M
  18. 18
    Claude 3.5 Sonnet
    Anthropic
    10.63
    31.9 / $3.00/M
  19. 19
    GPT-4o mini
    OpenAI
    0.00
    0.0 / $0.15/M
Open full leaderboard

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