Released: 2026-02-23
Key:
Score Threshold:
openai/gpt-4o
openai
((75 + 8) * 7) - 100
= 481
google/gemini-2.0-flash-001
(75 * 7) - 10 + 4
= 519
mistralai/mistral-medium-3
mistral
(100 + 75) * (8 - (10 - (7 - 4)))
= 175
google/gemini-3-flash-preview
(4 * 100) + 75 - 10
= 465
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (10 - (8 - 4))) + 75
= 675
x-ai/grok-3-mini-beta
x-ai
(100 * 4) + (75 - 10)
= 465
minimax/minimax-m2.5
minimax
100 * 4 + 75 - 10
= 465
z-ai/glm-5
z-ai
(100 * 4) + 75 - 10
= 465
moonshotai/kimi-k2.5
moonshot
100 * 4 + 75 - 10
= 465
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.