Released: 2026-04-16
Key:
Score Threshold:
openai/gpt-4o
openai
((75 + 4) * 5) + 100 + 25
= 520
google/gemini-3-flash-preview
(5 * 100) - (75 / (50 + 25))
= 499
deepseek/deepseek-chat-v3-0324
deepseek
((75 * (4 + 5)) - 50) + 100
= 725
z-ai/glm-5
z-ai
100 * 5 - (75 - 50) / 25
= 499
moonshotai/kimi-k2.5
moonshot
100*5-(4-75/25)
= 499
google/gemini-2.0-flash-001
(50 * 5) + 75 - 100 + 4
= 229
anthropic/claude-sonnet-4.6
anthropic
5 * 100 - 4 + 75 / 25
= 499
x-ai/grok-3-mini-beta
x-ai
(100 * 5) - (50 / 25)
= 498
minimax/minimax-m2.5
minimax
100 * 5 + 75 / 25 - 4
= 499
mistralai/mistral-medium-3
mistral
((100 + 75) * 4) + (25 - (50 / 5))
= 715
anthropic/claude-3.7-sonnet
anthropic
(100 * 5) - (75 / 25) + 4
= 501
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.