Released: 2026-02-19
Key:
Score Threshold:
openai/gpt-4o
openai
(25 * 10) - (50 / 5) + 7
= 247
google/gemini-2.0-flash-001
(50 * 5) - 10 + 7 - 25
= 222
minimax/minimax-m2.5
minimax
(25 + 7) * 10
= 320
deepseek/deepseek-chat-v3-0324
deepseek
((50 - 10) * 7) + (25 * 5)
= 405
anthropic/claude-3.7-sonnet
anthropic
25 * 10 - 50 + 7
= 207
x-ai/grok-3-mini-beta
x-ai
((50 * 5) - (25 - 7))
= 232
z-ai/glm-5
z-ai
50 * 5 - (25 - 7)
= 232
google/gemini-3-flash-preview
(5 * 50) - (25 - 7)
= 232
moonshotai/kimi-k2.5
moonshot
(25 - 7) * 10 + 50 + 10 / 5
= 232
mistralai/mistral-medium-3
mistral
(25 + 7) * (10 + 10) - 50
= 590
google/gemini-3.1-pro-preview
(((25 + (10 / 10)) * 7) + 50)
= 232
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.