AMA

Scenario guide

Best AI models for Data / BI Analyst

A spreadsheet / SQL / BI copilot that reads tables, answers analytical questions, and produces clean structured output. Weights a contamination-controlled data-analysis score heavily, plus long-context for large tables and instruction-following for rigid output formats.

Rankings use the same scenario weights and cost blending as the interactive leaderboard on AI Model Analyzer. Data is min-max normalised per benchmark; missing scores are skipped without penalty.

  1. 1
    Gemini 2.0 Flash
    Google
    Score 94.1Q 95.3In $0.10/M
  2. 2
    Gemini 1.5 Flash
    Google
    Score 86.7Q 83.7In $0.08/M
  3. 3
    Gemini 2.5 Pro
    Google
    Score 85.6Q 100.0In $1.25/M
  4. 4
    Llama 4 Maverick
    Meta
    Score 84.6Q 88.4In $0.27/M
  5. 5
    Gemini 1.5 Pro
    Google
    Score 84.6Q 97.7In $1.25/M
  6. 6
    Claude Sonnet 4 (Thinking)
    Anthropic
    Score 82.2Q 100.0In $3.00/M
  7. 7
    DeepSeek R1
    DeepSeek
    Score 77.8Q 83.7In $0.55/M
  8. 8
    Qwen3 235B (Thinking)
    Alibaba (Qwen)
    Score 77.3Q 76.8In $0.20/M
  9. 9
    Grok 3
    xAI
    Score 77.0Q 93.0In $3.00/M
  10. 10
    o3
    OpenAI
    Score 75.3Q 97.7In $10.00/M
  11. 11
    o3-mini
    OpenAI
    Score 74.7Q 83.7In $1.10/M
  12. 12
    Llama 4 Scout
    Meta
    Score 74.2Q 72.1In $0.18/M
  13. 13
    Claude 3.5 Sonnet
    Anthropic
    Score 73.5Q 88.4In $3.00/M
  14. 14
    GPT-4o mini
    OpenAI
    Score 73.1Q 69.8In $0.15/M
  15. 15
    GPT-4o
    OpenAI
    Score 71.0Q 83.7In $2.50/M
Open interactive leaderboard Build custom weightsHome