Released: 2026-02-20
Key:
Score Threshold:
openai/gpt-4o
openai
((75 + 25) * (6 + 2)) - 50
= 750
anthropic/claude-3.7-sonnet
anthropic
(75 * 100) / 25 * 6 * 2
= 3600
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 50) * 6) + (75 - (25 + 2))
= 948
x-ai/grok-3-mini-beta
x-ai
((100 + 50) * 6) + (75 / 25)
= 903
z-ai/glm-5
z-ai
(100 + 50) * 6 + 75 / 25
= 903
moonshotai/kimi-k2.5
moonshot
(100 + 50) * 6 + 75 / 25
= 903
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.