Released: 2026-02-14
Key:
Score Threshold:
google/gemini-2.0-flash-001
(100 * 10) - (50 + 10 + 2)
= 938
openai/gpt-4o
openai
((10 + 2) * 50) + 100 - 10
= 690
anthropic/claude-3.7-sonnet
anthropic
((100 - 2) * 10) - ((50 + 10) / 25)
= 977
mistralai/mistral-medium-3
mistral
(100 - (25 - 10)) * (50 - (10 / 2))
= 3825
deepseek/deepseek-chat-v3-0324
deepseek
(100 * (10 - 2)) + (50 - (25 - 10))
= 835
x-ai/grok-3-mini-beta
x-ai
((50 + 25) * 10) - (10 + 2)
= 738
qwen/qwen3-235b-a22b
qwen
((100 + 25 - 2) * ((50 + 10) / 10))
= 738
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.