Released: 2026-05-06
Key:
Score Threshold:
anthropic/claude-3.7-sonnet
anthropic
25 * (9 * 2 + 6 + 5 + 1)
= 750
anthropic/claude-sonnet-4.6
anthropic
25 * (9 * 6 - 5) + 2 * 1
= 1227
mistralai/mistral-medium-3
mistral
((25 * (9 + 5)) * (6 + 1)) + 2
= 2452
z-ai/glm-5
z-ai
((25 - 1) * (9 + 6) * 2) + 5
= 725
google/gemini-2.0-flash-001
((25 * 2) + 6) * (9 + 1)
= 560
deepseek/deepseek-chat-v3-0324
deepseek
((25 * (9 + 1)) + 6) * 5
= 1280
x-ai/grok-3-mini-beta
x-ai
25 * ((6 * 5) - (2 + 1))
= 675
google/gemini-3-flash-preview
((25 * 5 + 2) * (6 - 1)) + 9
= 644
openai/gpt-4o
openai
((25 * 9) - (6 / 2)) + 5
= 227
moonshotai/kimi-k2.5
moonshot
(25 + 6) * (9 * 2 + 5 - 1)
= 682
minimax/minimax-m2.5
minimax
(9 * 2 + 5 - 1) * (25 + 6)
= 682
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.