Released: 2026-07-18
Key:
Score Threshold:
openai/gpt-4o
openai
((25 * 10) + (3 * 3)) - 1
= 258
deepseek/deepseek-chat-v3-0324
deepseek
(25 * (10 - 3)) + (4 * (3 + 1))
= 191
minimax/minimax-m2.5
minimax
(25 - 1) * 10 - 3 * 4 - 3
= 225
moonshotai/kimi-k2.5
moonshot
3 * (25 * 3 - 4)
= 213
google/gemini-3-flash-preview
((25 - 4) * 10) + 3
= 213
mistralai/mistral-medium-3
mistral
((25 * (10 - 3)) * 4) + (3 * 1)
= 703
anthropic/claude-sonnet-4.6
anthropic
25 * (10 - 1) - 3 * 4
= 213
z-ai/glm-5
z-ai
25 * (10 - 1) - 4 * 3
= 213
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.