Released: 2026-03-27
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
(100 + 75) * (10 - (8 - 6))
= 1400
google/gemini-2.0-flash-001
(100 * 5) - 75 + 10 - 6 - 8
= 421
anthropic/claude-3.7-sonnet
anthropic
(100 * 5) - 75 - (10 - 8 - 6)
= 429
anthropic/claude-sonnet-4.6
anthropic
75 * 5 - 100 * (10 - 8 - 6)
= 775
deepseek/deepseek-chat-v3-0324
deepseek
((100 + 6) * (10 + 5)) - (75 + 8)
= 1507
z-ai/glm-5
z-ai
(75 - 5 * 8) * 10
= 350
minimax/minimax-m2.5
minimax
(100 - 75 + 6) * 10
= 310
moonshotai/kimi-k2.5
moonshot
(100 + 75) * (8 - 6)
= 350
openai/gpt-4o
openai
(75 * 5) - ((10 - 6) * 8)
= 343
google/gemini-3-flash-preview
(100 + 75) * (10 / 5)
= 350
x-ai/grok-3-mini-beta
x-ai
((100 + 75) * (10 / 5))
= 350
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.