Released: 2026-02-28
Key:
Score Threshold:
openai/gpt-4o
openai
((100 * 9) + 7) + (25 * 4)
= 1007
google/gemini-2.0-flash-001
(100 * 10) - 25 + 4
= 979
google/gemini-3-flash-preview
10 * (100 - (7 - 4)) + 9
= 979
anthropic/claude-3.7-sonnet
anthropic
(100 - 4) * 10 - 7
= 953
deepseek/deepseek-chat-v3-0324
deepseek
((100 * 9) + 25) + (10 * 4)
= 965
mistralai/mistral-medium-3
mistral
((100 + 7) * (4 + 9)) + (25 - 10)
= 1406
anthropic/claude-sonnet-4.6
anthropic
((100 - 9) * 10) - 25 + 4
= 889
x-ai/grok-3-mini-beta
x-ai
((100 * 10) - (25 - 4))
= 979
minimax/minimax-m2.5
minimax
(100*10)-(25-4)
= 979
z-ai/glm-5
z-ai
100 * 10 - 25 + 4
= 979
moonshotai/kimi-k2.5
moonshot
100 * 10 - (25 - 4)
= 979
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.