Released: 2026-04-23
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
(25 * 5) - (10 / (10 - (10 / 2)))
= 123
moonshotai/kimi-k2.5
moonshot
(10 * 10) + (25 - 5) - (2 / 10)
= 119
anthropic/claude-3.7-sonnet
anthropic
10 * 10 + 25 - 5 - 2 + 10
= 128
anthropic/claude-sonnet-4.6
anthropic
((25 * 5) - 10) * 2 - 10 + 10
= 230
deepseek/deepseek-chat-v3-0324
deepseek
(25 * 5) - (10 + 10 - 10 - 2)
= 117
z-ai/glm-5
z-ai
25 * 5 - (10 / 2 + 10 / 10)
= 119
minimax/minimax-m2.5
minimax
25 * 5 + 10/10 - 10
= 116
x-ai/grok-3-mini-beta
x-ai
(25 * 5) - (10 / 2) - (10 / 10)
= 119
openai/gpt-4o
openai
((25 * 5) - 10) - (10 / 2)
= 110
google/gemini-2.0-flash-001
(10 + 2) * 10 - 5
= 115
google/gemini-3-flash-preview
(10 * 10) + 25 - (10 / 2) + 5
= 125
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.