Released: 2026-04-26
Key:
Score Threshold:
google/gemini-2.0-flash-001
(100 * 8) + 75 + 4 - 50
= 829
openai/gpt-4o
openai
((100 * 8) + (75 * 4) - 4)
= 1096
deepseek/deepseek-chat-v3-0324
deepseek
(100 * 8) + (75 * 4) + (50 + 4)
= 1154
x-ai/grok-3-mini-beta
x-ai
((100 * 8) + 75 - 4 - 4)
= 867
minimax/minimax-m2.5
minimax
(100 + 75 + 50) * 4 - 8 * 4
= 868
anthropic/claude-3.7-sonnet
anthropic
100 * 8 + 75 - 4 - 4 + 50
= 917
anthropic/claude-sonnet-4.6
anthropic
100 * 8 + 75 - 4 - (50 / 4)
= 858
mistralai/mistral-medium-3
mistral
((100 + 75) * 8) - (50 - (4 / 4))
= 1351
z-ai/glm-5
z-ai
(100 + 75 + 50 - 8) * 4
= 868
moonshotai/kimi-k2.5
moonshot
4*(100+75+50) - 8*4
= 868
google/gemini-3-flash-preview
((100 + 8) * 4) * (75 / 50) + 4
= 652
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.