Released: 2026-04-14
Key:
Score Threshold:
google/gemini-2.0-flash-001
(100 * 4) + 75 + 10 + 9 - 25
= 469
openai/gpt-4o
openai
(75 + 9) * 4 + 25 + 10
= 371
deepseek/deepseek-chat-v3-0324
deepseek
(100 + 75) * (10 - 4) + (25 + 9)
= 1084
minimax/minimax-m2.5
minimax
(75 * 9) + 25 - (100 / 4)
= 675
moonshotai/kimi-k2.5
moonshot
(100 - 9) * (10 - 4) - 75 / 25
= 543
google/gemini-3-flash-preview
(9 * (100 - 25)) / (10 / 4) + 75
= 345
z-ai/glm-5
z-ai
(75 + 4) * 9 - 100 - 10 + 25
= 626
x-ai/grok-3-mini-beta
x-ai
9 * (75 - 25 + 10)
= 540
mistralai/mistral-medium-3
mistral
(100 * (4 + 9)) - (75 / 25) + 10
= 1307
anthropic/claude-3.7-sonnet
anthropic
(9 * 10 * 75) / 25 + 100 + 4
= 374
anthropic/claude-sonnet-4.6
anthropic
75 * 9 - 100 - 25 + (10 - 4)
= 556
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.