Released: 2026-06-17
Key:
Score Threshold:
openai/gpt-4o
openai
(4 * 25) + (6 * 75) + 50 - 100
= 500
mistralai/mistral-medium-3
mistral
((75 + 25) * (100 - (6 * 4)))
= 7600
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 6) * (75 - 25)) - (50 * 4)
= 5100
moonshotai/kimi-k2.5
moonshot
(100 + 4) * (6 + 50 / 25)
= 832
minimax/minimax-m2.5
minimax
(75+25)*6+100+50+4
= 754
google/gemini-3-flash-preview
(100 + 4) * (25 + 6 + 50 - 75)
= 624
z-ai/glm-5
z-ai
((100 / 50) + 6) * (75 + 25 + 4)
= 832
anthropic/claude-sonnet-4.6
anthropic
((25 + 75) * 4) + (100 * 6) * 50
= 30400
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.