Released: 2026-06-27
Key:
Score Threshold:
openai/gpt-4o
openai
(6 * 100) + 50 + 25 - 5
= 670
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 75) * 5) + 50 + 25
= 950
mistralai/mistral-medium-3
mistral
(100 - (75 - (50 + 25))) * (6 + 5)
= 1100
moonshotai/kimi-k2.5
moonshot
50 * (25 - 6) + (100 / 5)
= 970
minimax/minimax-m2.5
minimax
50 * (100 - 75 - 6) + (25 - 5)
= 970
z-ai/glm-5
z-ai
50 * (25 - 6) + 100 / 5
= 970
google/gemini-3-flash-preview
(100 * (5 + 50)) / 6 + 25 + 75
= 1016
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.