Released: 2026-08-21
Key:
Score Threshold:
openai/gpt-4o
openai
((75 + 5) * 10) - (3 + 4)
= 793
mistralai/mistral-medium-3
mistral
(75 * (10 - 3)) - (50 / (5 - 4))
= 475
google/gemini-3-flash-preview
((75 + 50) * 5) + 10 + 4
= 639
minimax/minimax-m2.5
minimax
(75 + 50) * 5 + (10 + 4)
= 639
deepseek/deepseek-chat-v3-0324
deepseek
(75 * (10 - 3)) + (50 * 4) - 5
= 720
moonshotai/kimi-k2.5
moonshot
50*(10+3) - 75/5 + 4
= 639
z-ai/glm-5
z-ai
(75 + 50) * 5 + 10 + 4
= 639
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.