Released: 2026-03-09
Key:
Score Threshold:
google/gemini-2.0-flash-001
(100 * 6) + 75 + 50 + 2 * 2
= 729
mistralai/mistral-medium-3
mistral
((75 * (100 - 6)) + 50) / 2
= 3550
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 2) * (6 + 2)) + 75
= 891
openai/gpt-4o
openai
((75 * 2) + 50) + 100 + 6 * 2
= 312
google/gemini-3-flash-preview
(100 - 2) * (2 + 75 / (50 / 6))
= 1078
x-ai/grok-3-mini-beta
x-ai
(( (100 + 75) * 6) - 50 - 2 - 2)
= 996
anthropic/claude-sonnet-4.6
anthropic
((75 + 2) * 100) / (50 / 6 + 2)
= 745
minimax/minimax-m2.5
minimax
75 * 6 * 2 + 100 - 50 + 2
= 952
anthropic/claude-3.7-sonnet
anthropic
(75 + 6) * 100 / (50 / 2 / 2)
= 648
z-ai/glm-5
z-ai
(100 + 2) * (50 + 2) / 6 + 75
= 959
moonshotai/kimi-k2.5
moonshot
75*6*2+100-50+2
= 952
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.