Released: 2026-04-15
Key:
Score Threshold:
openai/gpt-4o
openai
100 + 75 - 8
= 167
google/gemini-2.0-flash-001
(75 / (2 + 6)) * 100 - 8
= 929
mistralai/mistral-medium-3
mistral
(100 + 75) - (25 * (8 - (6 / 2)))
= 50
google/gemini-3-flash-preview
100 + 75 - 8
= 167
minimax/minimax-m2.5
minimax
100 + 75 - 8
= 167
deepseek/deepseek-chat-v3-0324
deepseek
(100 - (8 - 2)) + 75
= 169
anthropic/claude-3.7-sonnet
anthropic
(25 * 6) + (8 * 2) + 75
= 241
x-ai/grok-3-mini-beta
x-ai
(100 + 75) - 8
= 167
z-ai/glm-5
z-ai
100 + 75 - 8
= 167
moonshotai/kimi-k2.5
moonshot
100 + 75 - 8
= 167
anthropic/claude-sonnet-4.6
anthropic
100 + 75 - 8 * (25 - 6 * 2)
= 71
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.