Released: 2026-05-05
Key:
Score Threshold:
google/gemini-2.0-flash-001
(75 * 50 - 100) / 3 + 1
= 1217
mistralai/mistral-medium-3
mistral
(100 + (25 * 75)) - (50 * (3 + 1))
= 1775
deepseek/deepseek-chat-v3-0324
deepseek
((75 - 1) * (100 - 50)) / 25 - 3
= 145
z-ai/glm-5
z-ai
3 * (75 + 50 + 100/25 + 1)
= 390
moonshotai/kimi-k2.5
moonshot
(75 + 50 + 100/25 + 1) * 3
= 390
openai/gpt-4o
openai
100 + 75 + 50 + 1 + 3 * 25
= 301
x-ai/grok-3-mini-beta
x-ai
((100 * 3) + 75 + 25)
= 400
anthropic/claude-3.7-sonnet
anthropic
(100 * 3) + (75 - 1) * 50
= 4000
minimax/minimax-m2.5
minimax
100 * (1 + 3) - (75 - 50 - 25)
= 400
google/gemini-3-flash-preview
(100*3)+(75+50/25)
= 377
anthropic/claude-sonnet-4.6
anthropic
75 * (3 + 1) + 100 + 25 - 50
= 375
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.