Released: 2026-08-01
Key:
Score Threshold:
deepseek/deepseek-chat-v3-0324
deepseek
(75 * (8 + (8 / (4 - 1)))) + 3
= 803
google/gemini-3-flash-preview
(8 * (75 + 8)) - (4 - 1 - 3)
= 664
mistralai/mistral-medium-3
mistral
((75 * (8 + 1)) * 8) + (4 / 3)
= 5401
openai/gpt-4o
openai
(8 * 75) + (8 * 4) + 3 + 1
= 636
z-ai/glm-5
z-ai
(75 + 8) * 8 - (4 - 3 + 1)
= 662
moonshotai/kimi-k2.5
moonshot
(75 + 8) * 8 - 4 / (3 - 1)
= 662
minimax/minimax-m2.5
minimax
(75 + 8) * 8 - 3 + 1
= 662
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.