Skip to content

Value ranking

Best value on GPQA Diamond

As of October 25, 2026, the best value on GPQA Diamond is Llama 3.3 70B Instruct, at 302.70 normalized points per dollar of input price ($0.10 per million tokens, score 30.3).

Graduate-level Google-proof Q&A in physics, chemistry, and biology. Diamond subset is the hardest tier with PhD-validated answers.

“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
    Llama 3.3 70B Instruct
    Meta
    302.70
    30.3 / $0.10/M
  2. 2
    Llama 3.1 70B Instruct
    Meta
    170.27
    68.1 / $0.40/M
  3. 3
    Qwen2.5 72B Instruct
    Alibaba (Qwen)
    166.67
    60.0 / $0.36/M
  4. 4
    Mistral Large 2
    Mistral
    50.00
    100.0 / $2.00/M
  5. 5
    Mixtral 8x22B
    Mistral
    29.46
    58.9 / $2.00/M
Open full leaderboard

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