Released: 2026-05-27
Key:
Score Threshold:
google/gemini-2.0-flash-001
(50 * 10) + 4 + 3 - 5 + 4
= 506
google/gemini-3-flash-preview
(50 * (10 + 3)) - (4 + 5)
= 641
openai/gpt-4o
openai
(50 * 10) + ((4 + 4 + 5) * 3)
= 539
deepseek/deepseek-chat-v3-0324
deepseek
(50 * (10 + 3)) + (5 * 4) + 4
= 674
mistralai/mistral-medium-3
mistral
((50 + 4) * (10 + 5)) + (4 * 3)
= 822
z-ai/glm-5
z-ai
50 * (10 + 3) - (4 + 5)
= 641
minimax/minimax-m2.5
minimax
50 * (10 + 3) - 5 - 4
= 641
moonshotai/kimi-k2.5
moonshot
50 * (10 + 3) - (5 + 4)
= 641
anthropic/claude-sonnet-4.6
anthropic
(50 + 4) * (3 + 10) - 5 - 4
= 693
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.