CookingBench

Leaderboard / categories

Substitutions

Ingredient swaps with correct ratios, including allergen-aware alternatives.

Ranking

  1. 1Grok 4.3
    99.4
  2. 2GPT-5.5
    95.8
  3. 3Claude Sonnet 4.6
    95.6
  4. 4Claude Opus 4.8
    94.2
  5. 5Claude Fable 5
    91.7
  6. 6GPT-5.4 Mini
    91.6
  7. 7DeepSeek V4 Pro
    91.6
  8. 8Kimi K2.6
    88.9
  9. 9Qwen 3.5 Plus
    88.7
  10. 10Mistral Large 3
    85.5
  11. 11Gemini 3.5 Flash
    79.8
  12. 12Llama 4 Maverick
    77.8
  13. 13Gemini 3.1 Pro Preview
    70.7

Question heatmap (public questions only)

Model001002003004005006007008009010011012013014015016017018019020021
Grok 4.3
GPT-5.5
Claude Sonnet 4.6
Claude Opus 4.8
Claude Fable 5
GPT-5.4 Mini
DeepSeek V4 Pro
Kimi K2.6
Qwen 3.5 Plus
Mistral Large 3
Gemini 3.5 Flash
Llama 4 Maverick
Gemini 3.1 Pro Preview

Each cell is one question; deeper colour = higher score. Hover for exact values.