Released: 2026-05-26
Key:
Score Threshold:
google/gemini-2.0-flash-001
(50 - 6) * (10 - 2) - 8 * 3
= 328
openai/gpt-4o
openai
((50 * 6) - 2) + (8 * 3)
= 322
mistralai/mistral-medium-3
mistral
((50 * (10 - 3)) - (8 / (6 - 2)))
= 348
deepseek/deepseek-chat-v3-0324
deepseek
(50 - (10 - (8 / (6 - 2)))) * 3
= 126
z-ai/glm-5
z-ai
6 * (50 - 10 - 2) + 3
= 231
google/gemini-3-flash-preview
((50 - 10 - 2) * 6) + 3
= 231
moonshotai/kimi-k2.5
moonshot
6 * (50 - 10) - (8 - 2 + 3)
= 231
minimax/minimax-m2.5
minimax
(50+8)*(10-6)-(3-2)
= 231
anthropic/claude-sonnet-4.6
anthropic
50 * (6 - 2) + 8 * (10 - 3)
= 256
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.