Released: 2026-07-03
Key:
Score Threshold:
openai/gpt-4o
openai
(4 * 50) + (100 / 10) + 7 - 3
= 214
deepseek/deepseek-chat-v3-0324
deepseek
(50 * 4) + (10 * 3) + 100 - 7
= 323
anthropic/claude-sonnet-4.6
anthropic
100 * 3 - 50 - 10 - 7 - 4
= 229
moonshotai/kimi-k2.5
moonshot
(100 * 3) - (50 + 7 + 10 + 4)
= 229
minimax/minimax-m2.5
minimax
100 * 3 - 50 - 10 - 7 - 4
= 229
z-ai/glm-5
z-ai
100 + 3 * (50 - 7)
= 229
mistralai/mistral-medium-3
mistral
(100 + 50) + (10 * (7 + (4 * 3)))
= 340
google/gemini-3-flash-preview
(100*3)-(50+(4*7)-10)
= 232
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.