Released: 2026-06-18
Key:
Score Threshold:
mistralai/mistral-medium-3
mistral
(100 + 50 + 3) * (9 + 1)
= 1530
google/gemini-3-flash-preview
(100 * 3) + 25 - (9 + 50 - 1)
= 267
deepseek/deepseek-chat-v3-0324
deepseek
(100 - (50 - (25 * (9 + 3))))
= 350
moonshotai/kimi-k2.5
moonshot
3 * 100 + 9 + 1 + 50 / 25
= 312
z-ai/glm-5
z-ai
(9 + 3) * (25 + 1)
= 312
openai/gpt-4o
openai
(9 - 1) * 50 - (100 - 25)
= 325
anthropic/claude-sonnet-4.6
anthropic
3 * 100 + 9 + 1 + 50 / 25
= 312
minimax/minimax-m2.5
minimax
100 * 3 + 9 + 1 + 50/25
= 312
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.