Released: 2026-05-15
Key:
Score Threshold:
google/gemini-3-flash-preview
((100 + 10 + 25 + 1) * 5) - 50
= 630
moonshotai/kimi-k2.5
moonshot
(25 * 5) + (100 - 50) - 1 + 10
= 184
minimax/minimax-m2.5
minimax
50 * (10 + 1) + 100 + 25 + 5
= 680
x-ai/grok-3-mini-beta
x-ai
100 * (5 + 1) + 50 + 25 + 10
= 685
google/gemini-2.0-flash-001
(50 * 10) + 100 + 25 + 5 + 1
= 631
z-ai/glm-5
z-ai
((25 - 1) * 5 * 10) - (100 + 50)
= 1050
openai/gpt-4o
openai
((100 + 50) * 5) + (25 + 10 - 1)
= 784
deepseek/deepseek-chat-v3-0324
deepseek
(100 - (50 - (25 - 10) - 1)) * 5
= 330
mistralai/mistral-medium-3
mistral
((100 + 50) * (5 + 1)) + (25 - 10)
= 915
anthropic/claude-sonnet-4.6
anthropic
100 * (5 + 1) + 50 * (25 - 10)
= 1350
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.