Released: 2026-07-28
Key:
Score Threshold:
openai/gpt-4o
openai
(50 * 4) + (10 * 5) - 1
= 249
deepseek/deepseek-chat-v3-0324
deepseek
50 * (5 - 1) - (10 + 4 + 2)
= 184
z-ai/glm-5
z-ai
((50 * 4) - (10 * 2)) - 1
= 179
minimax/minimax-m2.5
minimax
(50 + 10) * (5 - 2) - 1
= 179
moonshotai/kimi-k2.5
moonshot
50 * 4 - 10 - 5 * 2 - 1
= 179
google/gemini-3-flash-preview
(50 * (4 - 1)) + (10 * (5 - 2))
= 180
mistralai/mistral-medium-3
mistral
((50 * 4) - (10 / (5 - 1))) + 2
= 199
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.