Released: 2026-03-07
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
(100 * (50 / (75 - 25))) + (3 * 1)
= 103
google/gemini-2.0-flash-001
(75 * (3 + 1)) + (100 - 50) + 25
= 375
openai/gpt-4o
openai
(75 * 3) + 100 - (50 / 25)
= 323
anthropic/claude-3.7-sonnet
anthropic
(100 * 3 + 50) * (25 - 1)
= 8400
deepseek/deepseek-chat-v3-0324
deepseek
((75 * (100 - 3)) - (50 * 1))
= 7225
x-ai/grok-3-mini-beta
x-ai
75 * (3 * (100 / 50) + 1)
= 525
moonshotai/kimi-k2.5
moonshot
(100 + 3) * (50 + 75) / 25
= 515
z-ai/glm-5
z-ai
(100 + 3) * ((75 + 50) / 25)
= 515
google/gemini-3-flash-preview
50 * (75 + 25) / 100 + 3 + 1
= 54
minimax/minimax-m2.5
minimax
(100 + 3) * ((75 + 50) / 25)
= 515
anthropic/claude-sonnet-4.6
anthropic
100 * 25 / 50 * (3 + 1)
= 200
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.