Released: 2026-06-04
Key:
Score Threshold:
openai/gpt-4o
openai
((50 * 9) - 7) - (25 - 6)
= 424
mistralai/mistral-medium-3
mistral
((50 + 25) * (9 + 7)) + (6 * 7)
= 1242
deepseek/deepseek-chat-v3-0324
deepseek
((25 * 7) - 6) * (50 - (9 + 7))
= 5746
minimax/minimax-m2.5
minimax
50*25 - 9*7*6
= 872
moonshotai/kimi-k2.5
moonshot
(9 * 7 - 50) * (25 + 6 * 7)
= 871
z-ai/glm-5
z-ai
(50 + 6) * (7 + 9) - 25
= 871
google/gemini-3-flash-preview
(50 + 7) * (9 + 6) + 25 - 7
= 873
anthropic/claude-sonnet-4.6
anthropic
( 50 + 7) * 9 * 6 / 25 + 7
= 130
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.