Released: 2026-05-31
Key:
Score Threshold:
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (9 - (7 / (5 + 3)))) * 6
= 4875
google/gemini-3-flash-preview
(100 + 3 + 9) * 7
= 784
mistralai/mistral-medium-3
mistral
(100 - (9 + 7)) * (6 + 5) + 3
= 927
openai/gpt-4o
openai
(100 + 7) * (6 + 3) - 5
= 958
google/gemini-2.0-flash-001
(100 - 6) * (9 - 5) * 3 + 7
= 1135
minimax/minimax-m2.5
minimax
100 * 7 + 6 * (9 + 5)
= 784
z-ai/glm-5
z-ai
(100 * 7) + (6 * (9 + 5))
= 784
anthropic/claude-sonnet-4.6
anthropic
((100 - 7) * 9) - 5 - 6 * 3
= 814
moonshotai/kimi-k2.5
moonshot
(100 + 9 + 3) * 7
= 784
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.