Released: 2026-03-06
Key:
Score Threshold:
deepseek/deepseek-chat-v3-0324
deepseek
(100 + 75) * (4 - (4 / 50))
= 686
google/gemini-2.0-flash-001
(25 * 100) / (75 - 4) + 4
= 39
mistralai/mistral-medium-3
mistral
(100 + 75 + 50) * 4 - 25
= 875
openai/gpt-4o
openai
(75 * 4) - ((100 + 50) / 4)
= 262
anthropic/claude-3.7-sonnet
anthropic
4 * (75 - 4)
= 284
minimax/minimax-m2.5
minimax
(50 * 4) + 75 + (100 / 25)
= 279
anthropic/claude-sonnet-4.6
anthropic
( 75 + 25 - 4) * 4 - 100 + 50
= 334
x-ai/grok-3-mini-beta
x-ai
(100 + (4 * 50)) - ((75 / 25) * 4)
= 288
z-ai/glm-5
z-ai
(75 + 4) * 4 - 25 - 100/50
= 289
moonshotai/kimi-k2.5
moonshot
(75 + 4) * 4 - 25 - 100 / 50
= 289
google/gemini-3-flash-preview
(75 * 4) - (50 / (25 / (100 / 4)))
= 250
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.