Released: 2026-03-22
Key:
Score Threshold:
openai/gpt-4o
openai
(100 * 9) - (7 * 6)
= 858
google/gemini-2.0-flash-001
(100 * 9) - 6 - 8 + 1 + 7
= 894
google/gemini-3-flash-preview
(100 + 1) * 9 - (7 + 6)
= 896
anthropic/claude-3.7-sonnet
anthropic
100 * 9 - (7 - 1) * 8
= 852
deepseek/deepseek-chat-v3-0324
deepseek
((100 * 9) + (8 * 7)) - 1
= 955
mistralai/mistral-medium-3
mistral
((100 - (8 + 1)) * (9 + 6)) - 7
= 1358
z-ai/glm-5
z-ai
(100 * 9) - (8 - 6)
= 898
moonshotai/kimi-k2.5
moonshot
(100 * 9) - (7 - 1) * (8 - 6)
= 888
x-ai/grok-3-mini-beta
x-ai
(100 * 9) - (8 - 6)
= 898
minimax/minimax-m2.5
minimax
100*9-(7-6+1)
= 898
anthropic/claude-sonnet-4.6
anthropic
( 9 * 100) - (7 - 6) + 1 - 8
= 892
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.