Released: 2026-04-28
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
((100 + 25) * (75 / 3)) + 1
= 3126
openai/gpt-4o
openai
((75 + 25) * 3) + (100 + 1)
= 401
google/gemini-3-flash-preview
((100 + 75 + 1) * 3) + 2 * 25
= 578
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (75 - (25 - 3))) - (2 * 1)
= 5298
x-ai/grok-3-mini-beta
x-ai
(75 * (3 * 2 + 1)) + (100 - 25)
= 600
minimax/minimax-m2.5
minimax
(100+75)*3+25*2+1
= 576
anthropic/claude-3.7-sonnet
anthropic
(3 * 75) + (100 * 2) + 25 + 1
= 451
moonshotai/kimi-k2.5
moonshot
(100 * 2 * 3 - 75 / 25 - 1)
= 596
z-ai/glm-5
z-ai
(100 - 2) * (75 / 25 + 3) - 1
= 587
anthropic/claude-sonnet-4.6
anthropic
(100 + 75 + 3) * 2 * (25 - 1)
= 8544
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.