Released: 2026-04-12
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
((100 - 5) * (75 - (50 / (8 + 3))))
= 6693
google/gemini-3-flash-preview
(50 * (100 + 75 + 5 + 3)) / 8
= 1143
anthropic/claude-3.7-sonnet
anthropic
(8 + 3) * 75 * 100 / 50
= 1650
openai/gpt-4o
openai
((75 + 5) * 8) + 100 - 3
= 737
deepseek/deepseek-chat-v3-0324
deepseek
(100 + 50) * 5 - (75 + 8 - 3)
= 670
x-ai/grok-3-mini-beta
x-ai
((75 - 8) * (50 / 5))
= 670
minimax/minimax-m2.5
minimax
(75+3)*8+100-50-5
= 669
moonshotai/kimi-k2.5
moonshot
(75 + 3) * 8 + 50 - 5
= 669
google/gemini-2.0-flash-001
(75 * 8) + 100 - 5 - 3
= 692
anthropic/claude-sonnet-4.6
anthropic
((100 + 75) * 3) + 8 * (50 - 5)
= 885
z-ai/glm-5
z-ai
(100 + 8) * 5 + 75 + 50 + 3
= 668
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.