Released: 2026-04-02
Key:
Score Threshold:
openai/gpt-4o
openai
100 + 10
= 110
google/gemini-3-flash-preview
100 + 10
= 110
deepseek/deepseek-chat-v3-0324
deepseek
100 + 10
= 110
anthropic/claude-3.7-sonnet
anthropic
100 + 10
= 110
z-ai/glm-5
z-ai
100 + 10
= 110
minimax/minimax-m2.5
minimax
100 + 10
= 110
anthropic/claude-sonnet-4.6
anthropic
100 + 10 + 6 - 25 + 75 - 50
= 116
x-ai/grok-3-mini-beta
x-ai
100 + 10
= 110
google/gemini-2.0-flash-001
100 + 10
= 110
moonshotai/kimi-k2.5
moonshot
100 + 10
= 110
mistralai/mistral-medium-3
mistral
(100 + (75 - (25 + 10))) * 6
= 840
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.