Released: 2026-04-07
Key:
Score Threshold:
anthropic/claude-3.7-sonnet
anthropic
100 * 7 + 25 * 3 + 9 + 3
= 787
openai/gpt-4o
openai
(100 * 7) + (3 * 25) + (3 * 9)
= 802
google/gemini-2.0-flash-001
(9 * (100 - 3)) - (25 + 7 + 3)
= 838
google/gemini-3-flash-preview
(3 * ((3 * 100) - (25 + 9 - 7)))
= 819
deepseek/deepseek-chat-v3-0324
deepseek
(100 - (25 - (9 - 3))) * (7 + 3)
= 810
anthropic/claude-sonnet-4.6
anthropic
( 100 * 7 ) + ( 25 * 3 ) + ( 9 - 3 )
= 781
x-ai/grok-3-mini-beta
x-ai
(100 * 7) + (25 * 3) + 9 + 3
= 787
moonshotai/kimi-k2.5
moonshot
3 * (7 * 25 + 100 - 9 - 3)
= 789
minimax/minimax-m2.5
minimax
(100 + 9) * 7 + 25 + 3/3
= 789
z-ai/glm-5
z-ai
(100 + 9) * 7 + 25 + 3 / 3
= 789
mistralai/mistral-medium-3
mistral
(100 + 25) * (9 + 3) + (7 * 3)
= 1521
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.