Released: 2026-07-05
Key:
Score Threshold:
openai/gpt-4o
openai
(100 + 75 + 9) * 2 + 2
= 370
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (10 - 2)) + (75 - (9 * 2))
= 857
moonshotai/kimi-k2.5
moonshot
9*75 - 100/2 - 10 + 2
= 617
minimax/minimax-m2.5
minimax
100 * (9 - 2) - 75 - 10 + 2
= 617
mistralai/mistral-medium-3
mistral
((100 + 75) * (10 - (9 - 2))) + 2
= 527
z-ai/glm-5
z-ai
75 * 9 - (100 / 2 + 10 - 2)
= 617
google/gemini-3-flash-preview
((75 - 10) * 9) + 100 / (2 + 2)
= 610
anthropic/claude-sonnet-4.6
anthropic
((100 + 9) * 2 - 75) * (10 - 2)
= 1144
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.