Released: 2026-06-29
Key:
Score Threshold:
openai/gpt-4o
openai
(4 * 100) + 75 + (2 * 25) + 1
= 526
mistralai/mistral-medium-3
mistral
((100 + 4) * (75 - 25)) - (2 * 1)
= 5198
deepseek/deepseek-chat-v3-0324
deepseek
(100 + (75 - 25)) * (4 + 1) - 2
= 748
minimax/minimax-m2.5
minimax
(100-75+4)*(25-2)-1
= 666
google/gemini-3-flash-preview
(75 - 1) * (100 - 25) / (2 * 4)
= 693
z-ai/glm-5
z-ai
(100 + 75) * 4 - 25 - 2 - 1
= 672
anthropic/claude-sonnet-4.6
anthropic
(75 + 100 + 25) * 4 - (2 - 1)
= 799
Each model receives the same prompt with the numbers to use. Models are tasked with creating an expression using only arithmetic operations to reach the target number. Each number can only be used once and you do not have to use all the numbers. Their answers are evaluated without feedback or retries.