Released: 2026-07-17
Key:
Score Threshold:
openai/gpt-4o
openai
((75 * 10) + (8 * 8)) + 3
= 817
mistralai/mistral-medium-3
mistral
(75 * (10 + 4)) - (8 * 8)
= 986
google/gemini-3-flash-preview
((75 + 3) * 10) + (8 * 8) + 4
= 848
deepseek/deepseek-chat-v3-0324
deepseek
((75 + 10) * (8 + 4)) - (8 - 3)
= 1015
moonshotai/kimi-k2.5
moonshot
10 * (75 + 8) + 8
= 838
anthropic/claude-sonnet-4.6
anthropic
(75 + 3) * 10 + 8 * (8 - 4)
= 812
z-ai/glm-5
z-ai
75 * 10 + 8 * (8 + 3)
= 838
minimax/minimax-m2.5
minimax
(75 + 8) * 10 + 8 - 4 - 3
= 831
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.