Released: 2026-04-29
Key:
Score Threshold:
google/gemini-2.0-flash-001
((50 * 7) + 75) - 6
= 419
mistralai/mistral-medium-3
mistral
((75 * (8 + 2)) * (7 - 50 / 6))
= -1000
openai/gpt-4o
openai
((75 * (8 + 6)) - (50 / 2)) + 7
= 1032
anthropic/claude-3.7-sonnet
anthropic
(75 + 50) * 8 - (2 * 6)
= 988
deepseek/deepseek-chat-v3-0324
deepseek
(75 * (8 + 6)) - (50 + 7)
= 993
moonshotai/kimi-k2.5
moonshot
75 * (8 + 7 - 2) - 6
= 969
x-ai/grok-3-mini-beta
x-ai
75 * (7 + 6) - (8 - 2)
= 969
z-ai/glm-5
z-ai
(75 * (8 + 7 - 2)) - 6
= 969
minimax/minimax-m2.5
minimax
75 * (7 + 6) + 2 - 8
= 969
anthropic/claude-sonnet-4.6
anthropic
75 * (8 + 6) - 50 - 7 + 2
= 995
google/gemini-3-flash-preview
7 * (50 + 75 + 6 + 8 - 2)
= 959
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.