Released: 2026-05-19
Key:
Score Threshold:
openai/gpt-4o
openai
((75 - 8) * 5) - 10
= 325
mistralai/mistral-medium-3
mistral
(100 + 75 + 50) * (8 - (10 - 5))
= 675
deepseek/deepseek-chat-v3-0324
deepseek
(75 - 50) * 10 + 100 - (8 + 5)
= 337
google/gemini-2.0-flash-001
(75 * 5) - (100 - 8)
= 283
google/gemini-3-flash-preview
((75 * 8) + 100) / (50 / 10) + 5
= 145
minimax/minimax-m2.5
minimax
(100 * 5) - (75 + 8 + 10)
= 407
z-ai/glm-5
z-ai
(50 + 8) * 5
= 290
anthropic/claude-sonnet-4.6
anthropic
100 * (75 - 50) / 10 + 5 * 8
= 290
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.