Released: 2026-06-07
Key:
Score Threshold:
openai/gpt-4o
openai
((75 - 2) * 10) + 3 + 25 + 100
= 858
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (10 - 3)) - (75 + 25 + 2)
= 598
mistralai/mistral-medium-3
mistral
(100 * (75 - (25 / (3 + 2)))) - 10
= 6990
z-ai/glm-5
z-ai
(100 + 75 + 2) * 3 + 25 + 10
= 566
google/gemini-3-flash-preview
(2*(3*(100-10)))+25
= 565
minimax/minimax-m2.5
minimax
(75 * 10) - (100 / 25) - 3 - 2
= 741
moonshotai/kimi-k2.5
moonshot
(100 * (10 - 2)) - (75 + 25 - 3)
= 703
anthropic/claude-sonnet-4.6
anthropic
((100+75)*3) + (25+10-2)
= 558
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.