AI vs Puzzles

Back to Home
Puzzle#155

Countdown

Released: 2026-03-31


11
Models Tested
3
Solved
9
LLM Providers

Key:

Used
Unused

Score Threshold:

Within 7

mistralai/mistral-medium-3

mistral

213Target
75
6
9
10
9
10

(75 * (10 - (9 / 9))) + (10 * 6)

= 735

Failed

google/gemini-2.0-flash-001

google

213Target
75
6
9
10
9
10

(10 * 10) + (9 + 75) - 6 - 9

= 169

Failed

deepseek/deepseek-chat-v3-0324

deepseek

213Target
75
6
9
10
9
10

(75 * (10 - 6)) + (9 + 9)

= 318

Failed

anthropic/claude-3.7-sonnet

anthropic

213Target
75
6
9
10
9
10

(9 + 6) * 10 * (10 - 9) + 75

= 225

Failed

MoonshotAI

moonshotai/kimi-k2.5

moonshot

213Target
75
6
9
10
9
10

(75 - 10/10) * (9 - 6) - 9

= 213

Perfect Solution!

google/gemini-3-flash-preview

google

213Target
75
6
9
10
9
10

(10 * (75 + 6)) / (10 + 9) + 9

= 51

Failed

anthropic/claude-sonnet-4.6

anthropic

213Target
75
6
9
10
9
10

75 * (10 - 6) + 9 * (10 - 9)

= 309

Failed

Minimax

minimax/minimax-m2.5

minimax

213Target
75
6
9
10
9
10

(10 + 10) * 6 + 75 + 9 + 9

= 213

Perfect Solution!

x-ai/grok-3-mini-beta

x-ai

213Target
75
6
9
10
9
10

( (75 * (9 - 6) ) - 10 )

= 215

Close; 2 away

Z.ai

z-ai/glm-5

z-ai

213Target
75
6
9
10
9
10

(10 + 10) * 6 + 75 + 9 + 9

= 213

Perfect Solution!

openai/gpt-4o

openai

213Target
75
6
9
10
9
10

(75 + 10 * 9) - 10 + 9

= 164

Failed

Methodology Note

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.

Leaderboard