Released: 2026-04-20
Key:
Score Threshold:
google/gemini-2.0-flash-001
(75 * 4) - (10 + 10) + 2 - 7
= 275
mistralai/mistral-medium-3
mistral
(75 * (10 - (7 - (10 / 2)))) + 4
= 604
openai/gpt-4o
openai
(75 * 4) - (10 - 2)
= 292
google/gemini-3-flash-preview
(10 * (75 / (7 - 4))) - 10 + 2
= 242
anthropic/claude-3.7-sonnet
anthropic
(75 - 2) * 4 - (7 - 10)
= 295
x-ai/grok-3-mini-beta
x-ai
((75 * 4) - (10 * 7) + 10 + 2)
= 242
moonshotai/kimi-k2.5
moonshot
(75 - 4) * 2 + 10 * 10
= 242
minimax/minimax-m2.5
minimax
75 * 4 - 7 * 10 + 10 + 2
= 242
deepseek/deepseek-chat-v3-0324
deepseek
((75 - (10 / (7 + 2))) * 4)
= 295
z-ai/glm-5
z-ai
(10 + 10 + 2) * (7 + 4)
= 242
anthropic/claude-sonnet-4.6
anthropic
75 * (2 + 10/10) + 4 * 7
= 253
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.