Released: 2026-06-21
Key:
Score Threshold:
google/gemini-3-flash-preview
(10 * (75 + 100 + 8 + 1 + 8))
= 1920
openai/gpt-4o
openai
(100 + 75 + (8 / 8) + 1) * 10
= 1770
mistralai/mistral-medium-3
mistral
((100 + 75) * (10 - (8 / 8))) + 1
= 1576
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 8) * 8) + (75 + 10 + 1)
= 950
anthropic/claude-sonnet-4.6
anthropic
75 * 10 + 100 + 8 + 8 - 1
= 865
z-ai/glm-5
z-ai
100 * 8 + 75 + 10 - 1
= 884
minimax/minimax-m2.5
minimax
100 * 8 + 75 + 10 - 1
= 884
moonshotai/kimi-k2.5
moonshot
100 * 8 + 75 + 10 - 1
= 884
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.