Released: 2026-02-25
Key:
Score Threshold:
openai/gpt-4o
openai
(75 * 10) + 7 + 50 + (100 / 25)
= 811
minimax/minimax-m2.5
minimax
(100 - 7) * 10 + 75
= 1005
google/gemini-3-flash-preview
(7 * (100 + 25)) - (50 + 75) / 10
= 862
deepseek/deepseek-chat-v3-0324
deepseek
(100 + 7) * (10 - (75 / 25))
= 749
anthropic/claude-3.7-sonnet
anthropic
((75 * 10) + 100) + (25 * 7 - 50)
= 975
x-ai/grok-3-mini-beta
x-ai
(100 * 7) + 75 + 50 + 10
= 835
moonshotai/kimi-k2.5
moonshot
7 * (75 + 50 - 10 + 100 / 25)
= 833
mistralai/mistral-medium-3
mistral
((100 + 7) * (75 - (50 - (25 / 10))))
= 2942
google/gemini-2.0-flash-001
(7 * (100 + 10)) + (75 - 50)
= 795
z-ai/glm-5
z-ai
7 * (75 + 50 - 10 + 100 / 25)
= 833
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.